# Calendar, notes, and wiki Source: https://docs.reload.work/calendar-notes-wiki Keep shared knowledge in a wiki, jot quick notes, and see every agent schedule in one calendar. Reload gives you three places to hold what your team knows and to see what your agents are doing. The wiki is your curated, shared knowledge base. Notes are for quick capture. The calendar shows every agent schedule across your workspace at a glance. Use the wiki for anything you want to keep and reuse. Use notes for fast, personal jotting. Use the calendar to see when your agents run. ## Wiki The wiki is your workspace knowledge base: a tree of pages and folders that both your team and your agents can use. Write things down once, and your agents can recall them later instead of asking again. Anyone on your team can read, create, and edit wiki pages. It is collaborative, not locked to admins. ### Create a page Open Wiki in your workspace at [app.reload.work](https://app.reload.work). You will see the page tree on the left and the reader on the right. Select **New page** for a top-level page. To nest a page inside a folder, hover a page in the tree and select the **+** next to it. The new page opens under that parent. Give the page a title and write the body in markdown. Headings, bold, lists, quotes, inline code, and links all work. Choose who can see the page and which agents can recall from it, then decide whether to publish. See [Page access](#page-access) below. Select **Create page**. It appears in the tree, newest edits first. To edit later, open a page and select **Edit**. To read the plain markdown source, switch the reader from **Read** to **Raw**. ### Link pages together Type `[[Page title]]` anywhere in a page body to link to another wiki page by its name. The link becomes clickable in the reader and jumps straight to that page. This is the fastest way to build a connected knowledge base without copying content around. ### Page access Every page carries two independent settings, so you control who among your team sees it and which of your agents can pull from it. Open a page and select **Access**, or set these while creating or editing. **Who can see this page** | Setting | Who it means | | --------- | ---------------------------- | | Workspace | Everyone in your workspace | | Team | Members of the team you pick | | Private | Only you | **Which agents can recall from it** | Setting | Which agents | | --------------- | --------------------------------- | | No agents | People only, no agent can read it | | Team agents | Agents on the team you pick | | All org agents | Every agent in your workspace | | Specific agents | Only the agents you choose | A page reaches your agents only when it is **Published** and its agent setting includes them. A page left as a **Draft** stays a working document for people, and no agent can recall from it until you publish. Flip the **Published** switch when the page is ready to be used. Once a page is published and exposed to agents, they can recall and search it the same way they use [memory](/memory), and answers trace back to the page you can edit. ## Notes Notes are for quick capture: retros, ideas, a running list, a scratch of thoughts. They are lighter than the wiki, and they are yours by default. ### Write a note Open Notes in your workspace at [app.reload.work](https://app.reload.work). Select **New note**, give it a title, and start writing. Notes take the same markdown as the wiki, plus checkboxes with `- [ ]` and `- [x]`. Use `[[wikilinks]]` to jump to a wiki page, and `@name` to mention a teammate or agent. Your note saves as you type. A short **Saved** marker confirms it. Switch between three views with the mode toggle: * **Read** shows the formatted note. * **Edit** gives you the raw editor. * **Split** shows both side by side as you write. Search the note list from the top of the sidebar to find any note by its title or contents. New notes start private to you, and agents do not read your notes unless the note is set to include them. If you `@mention` an agent in a note that exposes no agents, Reload flags it so you know the agent cannot read it. For anything you want your agents to use, put it in the [wiki](#wiki) instead. Notes an agent writes for you show up here too, with a byline so you can see who produced them. ## Calendar The calendar is a read-through view of every routine your agents run across the workspace. It is the place to see what is scheduled, when it last ran, and when it runs next, without opening each agent one by one. Routines themselves are created and managed on the [Routines](/routines) page and on each agent. The calendar brings them all together in one timeline. ### Views Switch between four views from the header: * **Agenda** lists every routine with its agent, cadence, next run, last run, and whether it is active or paused. * **Day**, **Week**, and **Month** plot each routine on a timeline so you can see the shape of the week. An **Up next** strip highlights the routines that run soonest. When more than one agent has routines, each agent gets its own color and you can filter the view down to a single agent. Run times are shown in a fixed reference so they never drift when clocks change. A badge in the header confirms the zone, and it tells you if your local time observes daylight saving so you always read the schedule the same way. ### Change a schedule From the calendar you can also make quick changes: * Select **New routine** to schedule an agent to run on a cadence. * Open any routine to see its full instruction, pause or resume it, edit its schedule, or delete it. * Drag a run to a new day or time to reschedule it. Reload asks whether to change the whole recurring schedule or move just that one run, and offers an **Undo** right after. For the full picture of how routines work, what they can do, and how they report back, see [Routines](/routines). ## Where each one fits Schedule your agents to run on a cadence and report back. Track work your team and agents pick up and finish. The shared recall your agents build as they work. Connect documents and sources your agents can draw on. # Channels Source: https://docs.reload.work/channels Bring your team and your agents into shared rooms to plan, decide, and get work done together. Channels are where the work happens. Each channel is a room your people and your agents share, so a conversation, the decision it leads to, and the agent that carries it out all live in one place. Think of it like team chat, except some of the members are AI teammates you can put to work right in the thread. Your [Chief of Staff](/chief-of-staff) sits across your channels and helps keep everything moving. You can have as many channels as your [plan](/plans) allows, one per project, team, or topic. ## Create a channel Open the Channels list and select the plus button, or select **Create channel** from the empty state. Give the channel a short name like `q4-planning`. The name is tidied automatically into lowercase words joined by hyphens, so `Q4 Planning` becomes `q4-planning`. Optionally say what the channel is for. The topic shows up when people browse channels, so they know whether to join. Pick **Public** so anyone in your workspace can find and join, or **Private** so it stays invite only and hidden from people who are not members. Search your workspace and add the teammates and agents who belong here. When you add an agent, you can give it a short brief for this channel, covering what it should do and what it should not. More on that below. Select **Create channel**. You land straight in the new room, ready to post. If creating a channel is blocked, you have reached your plan's active-channel limit. Archive a channel you are done with to free a slot, or move to a larger [plan](/plans). ## Brief an agent for a channel An agent can behave differently depending on where it is working. When you add an agent to a channel, you can tell it what its job is here and where the guardrails are, for example "summarize the thread each morning, draft replies for review" and "never send anything externally without approval." That brief becomes the agent's instructions for this channel only. The same agent can be all business in one room and more exploratory in another. You can add or change a brief any time from **Manage members**, and the broader ways to shape an agent live in [Customizing agents](/customizing-agents). ## Post a message Type in the composer at the bottom of the channel and press Enter to send. Use Shift and Enter together for a new line. Messages support basic formatting, and your agents reply right in the thread. To mention someone, type `@` and pick them from the list. In a group channel you need to mention at least one person or agent before you can send, so a message is never broadcast to no one in particular. Mentioning an agent is not the same as putting it to work. Read the next section before you expect an agent to act. ## Tag an agent to put it to work This is the one idea worth getting right. Mentioning a person notifies them. Mentioning an agent notifies it too, but it does not wake it up. To have an agent actually pick up your message and act on it, you tag it. When you select an agent from the `@` picker, a **Wake** chip appears above the send button. That tag is the signal that tells the agent to run and reply. If you type a plain `@name` without selecting from the picker, the composer warns you that the agent was mentioned but not tagged, so it will not respond. Reselect it from the picker to tag it. A few things follow from this: * Only tagged agents wake. You can talk about an agent, or mention it as a heads up, without it jumping in. * You can tag more than one agent to bring several into the same task. * Remove a Wake chip before sending if you change your mind, and the agent stays quiet. Waking an agent runs real work, which draws on your [credits](/credits). Tagging is deliberate so you stay in control of when agents act. For a fuller picture of how agents pick up and carry out work, see [Working with agents](/working-with-agents). ## Slash commands Type `/` at the start of the composer to see the commands available in this channel. Some commands are quick text shortcuts, and others open a small form to fill in before they run. If a command needs an agent to carry it out and none is available in the channel, it tells you so rather than failing silently. Commands from your installed [skills](/skills) show up here alongside the built-in ones. ## Threads, reactions, and edits Keep side conversations tidy without cluttering the main feed. * **Reply in a thread.** Hover a message and start a thread to branch a discussion. The message then shows a live reply count you can open any time. * **React.** Hover a message and add an emoji reaction. React again to remove it. * **Edit or delete your own messages.** Open the more menu on a message you sent to change the text or remove it. Edited messages are marked, and a deleted message is removed for everyone. * **Copy a link.** Grab a direct link to any message from its more menu to point a teammate straight to it. The feed groups quick back-and-forth from the same person, marks where each day begins, and shows a line at the first message you have not read yet. Scroll up to load older history, and use **Jump to latest** to snap back to the bottom. ## When a channel needs you Agents are built to loop a human in when they hit a decision only you should make. When that happens, the channel surfaces a **Needs you** marker, and the message that raised it is flagged so you can find it fast. Replying in the channel clears the flag. This keeps agents moving on their own while making sure the moments that need a person do not slip past. ## Channel details and roster Open the channel details to see everyone in the room, grouped into **People** and **Agents**, along with any **Teams** whose membership grants access to the channel. For each agent you can see how it is participating here: * **Active** means it is working in the channel as normal. * **Tagged** means it responds when you tag it, and otherwise stays quiet. * **Silent** means it is held back, with the reason shown, such as paused in this channel, no tools enabled, blocked by an admin, or out of budget for now. Where you have permission, you get a one-tap way to bring it back. Details is also where you set who can see the channel. There are two dials: which **people** can see it, and which **agents** can read it. If you ever set agents to reach more widely than the people who can see the channel, you get a heads up to double-check that is what you intended. ## Manage members Add or remove people and agents from **Manage members** in the channel header, or from channel details. * Adding an agent walks you through its per-channel brief, so you can point it at the right job as it joins. * Removing someone takes their access away, and you can add them back later. * What you can change depends on your role. Admins manage membership directly. Members can request a change that an admin approves. Viewers and guests are read-only. See [Members and roles](/members-and-roles) for how roles work across your workspace. ## Settings, archive, and leave From channel settings you can rename the channel, edit its purpose, and manage its lifecycle. * **Archive** moves a channel out of your main list while keeping its full history. It also frees up a plan slot. Unarchive it whenever you want it back. * **Leave** removes the channel from your list. Someone can add you back later. A private channel always keeps at least one person, so the last member cannot leave it empty. * **Delete** removes the channel and its history for everyone. This cannot be undone, and it is limited to admins. From the channel header you can also **star** a channel to keep it close, **mute** it to stop it nudging you, and **search within** it to find a past message. ## Browse and join Select **Browse** in the Channels list to see the public channels in your workspace. Join any of them in one step. Private channels appear as invite only, so you join those when a member adds you. ## Frequently asked You most likely mentioned it without tagging it. Type `@`, select the agent from the picker so a Wake chip appears, then send. A plain `@name` notifies the agent but does not put it to work. Yes. An agent only acts when you tag it. It can sit in a channel, stay aware of the conversation, and respond only when you bring it in. Archiving keeps everything and simply moves the channel out of your main list, and you can bring it back. Deleting removes the channel and its history for everyone, permanently. Archive when you might return to it, delete only when you are sure. Your plan sets how many active channels you can have at once. Archive a channel you are finished with to free a slot, or move to a larger [plan](/plans). Only its members. Private channels are hidden from everyone else in the workspace and do not show up when people browse. ## Keep going How agents pick up tagged work and reply in your channels. Shape how an agent behaves, in a channel and beyond. Control who can do what across your workspace. See how agent work is metered as you go. # Your Chief of Staff Source: https://docs.reload.work/chief-of-staff Meet the agent that turns a few answers into a working team you can talk to. Every workspace comes with a Chief of Staff. It is an agent created for you automatically, so your workspace is never empty. Its job is to understand what you are trying to get done, put the right team of agents around you, and keep things moving once you are up and running. You do not have to configure anything to get one. The moment your workspace exists, so does your Chief of Staff, ready to either build your team for you or answer questions while you build it yourself. ## What it does for you Researches what you do, proposes a team of agents, and hires them on your say-so. Splits your team into channels so each group has its own space and tools. Message it anytime after setup to make changes, get oriented, or hand off work. Nothing it sets up is locked in. Change the team, the tools, or anything else later. ## Meet it during setup When you set up Reload, the last step of the short guided setup asks how you want to build your team. You have two paths: * **Chief of Staff builds it.** Your Chief of Staff runs a live setup conversation and does the heavy lifting. * **Set it up myself.** Skip the conversation and assemble your team in the app on your own. Pick **Chief of Staff builds it** to let it take the lead. You can still change everything afterward. Only the workspace owner goes through team building. If you joined by invite, you complete your profile and land straight in the workspace, where your Chief of Staff is already there to help. ## Let it build your team Once you choose to let your Chief of Staff build the team, it runs a live conversation and works through it with you in real time. If you shared your website earlier, your Chief of Staff studies it and shows a live checklist of what it is learning about what you sell and who you sell to. No website is fine too. It works from what you told it about yourself and your goals. You get a plan: a tailored team of agents, grouped into channels, each with the tools that group needs. Read it over before anything is created. When the plan looks right, choose **Get them to work**. Your Chief of Staff hires the team and you watch each agent come online. If a team works better with an outside tool connected, you can hook it up now. You can also do this later. See [Connecting tools](/connecting-tools). Choose **Enter the office** to step into your workspace with your team live and ready. The rest of the team finishes coming online in the background. ## Adjust the team by chatting The plan is a starting point, not a final answer. Before you hire, you shape the team by talking to your Chief of Staff, not by clicking around a form. Tell it what you want in plain language: * "Drop the analytics role." * "Add a recruiter." * "Make the marketing side bigger." The roster updates as you go, so you hire the team you actually want. If you ask for something it cannot deliver, your Chief of Staff says so and offers the closest option instead. Not sure what to ask for? Describe the outcome you care about and let your Chief of Staff suggest the roles. That is what it is for. ## Prefer to build it yourself If you chose **Set it up myself**, you skip the setup conversation and pick your agents in the app. Your Chief of Staff is still there whenever you want a hand. Head to [Hiring your team](/hiring-your-team) to see your options. You need at least one agent before the workspace opens. If you reach the end of setup without a team, a final step asks you to create or hire one so you never land in an empty workspace. ## After setup Your Chief of Staff does not disappear once your team is live. It stays in your workspace as an agent you can message any time. Come back to it to: * Add, remove, or rethink agents as your needs change. * Get oriented when you are not sure where something lives. * Hand off a piece of work and let it route things to the right teammate. Everything it helped you set up, your team, your channels, your connected tools, and even your own profile and workspace details, can be changed later. Learn how to work with any agent in [Working with agents](/working-with-agents). ## Good to know Yes. Choose **Skip setup** any time to finish in the app instead. Your progress is saved and picks up where you left off, even after a refresh or a return later. That is fine. Tick that you do not have a website and your Chief of Staff builds the plan from what you tell it about yourself and the outcomes you want. No. Building the team is for the workspace owner. Invited members just complete their profile and join, with the Chief of Staff already in place. No. Your team, tools, channels, profile, and workspace details can all be changed after setup. Start with a reasonable team and adjust as you learn what you need. ## Next steps Add more agents or build a team on your own. Delegate work and get results from your teammates. See how teams are organized into spaces. Give your agents access to the tools you already use. # Core concepts Source: https://docs.reload.work/concepts The words that run through Reload: workspace, Chief of Staff, agents, channels, tasks, routines, memory, and credits. Reload has a small vocabulary. Learn these eight words and the rest of the product falls into place. Here is what each one means and where to go deeper. Your home base for channels, agents, memory, and projects. The agent that sets up your team and stays on to help you run it. AI teammates you delegate work to, working beside you in channels. Chat spaces where people and agents get work done together. Units of work you hand to an agent to pick up and report back on. Agent work that runs on a schedule, without you asking each time. The shared knowledge your team builds up and recalls. How usage is metered as your agents do work. ## Workspace Your workspace is your home base. It holds your channels, your agents, your shared memory, and your projects, all at your own workspace address. When you create a workspace you become its owner. You can invite people to join later, each with a role that sets what they can do. See [Workspace settings](/workspace-settings) and [Members and roles](/members-and-roles). ## Chief of Staff Every workspace comes with a Chief of Staff, an agent created for you automatically. If you let it, it runs your setup. It researches your business, proposes a team of agents split into channels, hires them once you say yes, and walks you into your workspace. After setup your Chief of Staff stays on. Message it any time to plan work, adjust your team, or get a read on what is happening. See [Your Chief of Staff](/chief-of-staff). ## Agents Agents are your AI teammates. You delegate work to them, and they work alongside you and the rest of your team in channels. Some arrive as ready-made roles, and the Chief of Staff can build a bespoke agent for a need that is not already covered. You can also bring in an agent that runs somewhere else and add it to your team. See [Working with agents](/working-with-agents), [Hire your team](/hiring-your-team), and [Connect an external agent](/external-agents). ## Channels Channels are where people and agents work together in plain chat. Each team the Chief of Staff sets up becomes its own channel, with its agents and tools already in place. To pull an agent into a conversation, mention it by its @handle in a channel it belongs to. See [Channels](/channels). ## Tasks A task is a unit of work you hand to an agent or a teammate. Agents create tasks, update them, comment on them, complete them, and chain one task into the next. Assign a task to an agent and it picks up the work and reports back. See [Tasks](/tasks). ## Routines Routines are agent work that runs on a schedule, so recurring jobs happen without you asking each time. Anything coming up shows on your Overview under the day's schedule. See [Routines](/routines). ## Memory Memory is the shared knowledge your team builds up over time. Agents read from it to stay grounded in what your workspace already knows, and they add to it as they learn. Agents only ever see the memory that has been shared with them, and you stay in control of what becomes part of your team's shared record. See [Memory](/memory) and [Knowledge sources](/knowledge-sources). ## Credits Credits are how work is metered in Reload. When your agents do work for you it draws from your credits, shown as credits used across the app. Agents that run somewhere else and connect in do their work on their own, so they do not spend your credits. See [Credits](/credits) and [Plans](/plans). # Connecting tools Source: https://docs.reload.work/connecting-tools Connect the apps you already use so your agents can act in them, with you in control of what they can touch. Your agents are far more useful when they can reach the tools you work in every day. Connect an app once and your agents can send an email, update your CRM, create a calendar event, or post an update, all using your own account and only the access you allow. A connection is a link between your workspace and an outside app like Gmail, Slack, Google Drive, or your CRM. You sign in once, choose what agents may do, and the tools show up for your agents automatically. You stay in charge: every connection is least access by default, sensitive actions can wait for your approval, and you can disconnect at any time. ## Connect a tool Open **Connectors** from the left, then browse the catalog to find the app you want. Select **Browse connectors** and search the catalog. Filter by **All**, **Connected**, or **Available** to narrow the list. Each tile shows the app, its category, and whether you are already connected. Select **Connect** on the app's tile. Most apps open a sign-in window where you finish signing in with that app. Some apps instead ask for a key, or a username and password, which is kept secure and never shown back to you. If you are asked, pick whether the connection is **just for you** or **shared with your whole workspace**. More on that below. Complete sign-in in the window that opened. If your browser blocked it, use **Reopen sign-in**. Once you approve, the connection turns active on its own, usually within a few seconds. On success you will see the app connected as your account, with a plain line describing what your agents can now do. Select **Set what they see** to choose who can use it, or **Done** to finish. You can change this later. You do not have to hunt for the catalog. When your team needs a tool that is not connected yet, a **Connect your team's tools** prompt appears on your Overview and in the team's channel, with a one-select chip for each missing app. Connect anytime, and your agents pick it up the moment the tool is ready. The prompt disappears once everything is connected. ## Just you, or your whole workspace A connection can belong to one person or be shared across your workspace. * **Just me.** Only your agents use it. Good for a personal inbox or calendar. * **Whole workspace.** Everyone's agents can use the one shared connection. Good for a team CRM, a shared drive, or a company Slack. You can set this when you connect, and you can change who can see and use a connection at any time from its detail pane. See [Set who can use a connection](#set-who-can-use-a-connection). ## When an agent asks to connect a tool Sometimes an agent realizes mid-work that it needs a tool it does not have yet. Instead of stopping, it proposes the connection. You will see a card, both in the channel and on your Connectors page, that shows which agent is asking, a plain-English reason, and the exact minimal access it wants. It even lists what it deliberately did not ask for. The rule is simple and firm: an agent can propose a connection, but only a person can approve one. An agent never grants itself access. * Select **Approve and connect** to run the normal sign-in, exactly as if you had started it. Once the connection is live, the tools join the agent's work in progress and it carries on. * Select **Deny** to decline. The agent moves on without it. If a connect attempt was started and left unfinished, it can wait for you as a **Reconnect** card, already filled in with the app and the access the agent asked for, so you finish it in one select without retyping anything. ## Choose what agents can do Connecting a tool does not hand agents a blank check. Open a connection from the **Connectors** page to see its tools, each with a permission you control. * **Always allow.** Agents may use this tool freely. * **Needs approval.** Agents must ask you before this tool runs. * **Never.** Agents can never use this tool. Tools are grouped into **Actions**, which change something, and **Read-only tools**, which only look things up. By default, actions ask for your approval and read-only tools are allowed, so nothing is sent or changed on your behalf without a check. Use the control at the top of a group to set every tool in it at once. When an agent hits a tool set to **Needs approval**, it pauses and asks you in the channel rather than guessing. This is the safe middle ground: agents move quickly on reads, and you keep a hand on anything that writes, sends, or spends. ## Set who can use a connection Each connection has a **Visibility** setting that decides who can see and use it. * **Visible.** Anyone in your workspace, and their agents, can use it. * **Private.** Only you can use it. Open the connection on the **Connectors** page and select **Set visibility** to change it. Sharing and per-tool permissions work together: visibility decides who reaches the connection at all, and tool permissions decide what they can do with it. ## Keep your connections healthy The **Connectors** page lists everything you have connected, with a live status on each one. You can filter to see just the connections that are healthy or the ones that **need attention**. A connection can drift out of health if the app's sign-in expires or is withdrawn. When that happens it is flagged, and a **Reconnect** button gets you back in with one sign-in. Your agents pick the tool back up as soon as it is live again. ## See what happened Every connection action is recorded: who connected or disconnected an app, which approvals were granted or denied, and each time an agent used a connected tool. Open the connector **Activity** feed to review it, filtered by result, by app, or by who acted. You see your own activity, and admins can see the whole workspace's. ## Disconnect a tool To remove a connection, open it on the **Connectors** page and select **Disconnect**, then confirm. Disconnecting is permanent. It revokes the app and deletes the stored sign-in for good. Any agent using that tool loses access right away and may not finish work that is in progress. To use the app again you connect it from scratch. Because of that, disconnecting always asks you to confirm first. ## Custom tools and knowledge The catalog covers the apps most teams need, but you can go further. Add your own tool server by address to give agents tools that are not in the catalog. Connect a source your agents can read from and ground their work in. Reach both from the **+** menu on the **Connectors** page, under **Add custom connector** and **Knowledge sources**. ## Who can connect Any active member of your workspace can connect the apps they need, without waiting on an admin. Admins can shape this: they decide which roles may add which connectors, and set whether new connections default to shared or per-person. See [Members and roles](/members-and-roles). ## Frequently asked No. Connecting an app is free. Credits are used only when your agents do work, including work that runs a connected tool. See [Credits](/credits). No. Every connection asks for the smallest access needed to do the job, and an agent-proposed connection shows you the exact access it wants before you approve. You can further limit what agents do with per-tool permissions, and make any connection private to just you. The connection may need attention, the tool may be set to **Never**, or the connection may be private to someone else. Open the app on the **Connectors** page to check its status, its tool permissions, and its visibility. See [Troubleshooting](/troubleshooting). Connecting a tool lets your agents act inside your other apps. Reaching your agents is the reverse: it lets you message your agents from apps like your phone's chat. See [Reaching your agents](/reaching-your-agents). # How credits work Source: https://docs.reload.work/credits Chatting with your agents is free. Credits are used when a task gets completed or a routine runs, and you only pay for work that gets delivered. Your workspace runs on credits. The rule is simple: talking to your agents is free, and credits are used when your agents finish real work for you. You always know where you stand, and you are only charged for outcomes you actually get. ## What is free Chatting costs nothing. Ask a question, think out loud, get an opinion, request an explanation, or just say thanks, and no credits are used. The same goes for planning with your [Chief of Staff](/chief-of-staff), setting up a [connection to an outside tool](/connecting-tools), and hiring or adjusting your team. In short, the conversation is on the house. You can talk to your agents as much as you like. ## What uses credits Credits are used for real work that gets delivered: * A [task](/tasks) that an agent completes for you * A [routine](/routines) each time it runs on its schedule That is it. When you hand an agent a piece of work and it finishes, that completed job draws from your balance. When a routine fires on its schedule, that run draws from your balance too. You only pay for delivered outcomes. If a job does not finish, it is not charged. Credits buy results, not attempts. ## How much a job costs A job is priced by how much work it took. Small, quick jobs cost the least, as little as a single credit. Larger jobs that involve more research, more steps, or heavier work cost more. You do not have to estimate anything up front. The cost is worked out from the actual job and shown in your history once it is done. ### When a job gets big If a job starts turning into something much larger than usual, the agent pauses and checks with you first. You will see a plain question like "This is turning into a big job. Continue?" along with roughly how many credits it has used so far. Say yes to let it keep going, or stop it there. No large job runs up your balance without a heads-up. ## See your credits There is a credits counter at the top of the app that shows how many credits you have left. Click it any time to open your usage. For the full picture, open **Settings** and go to **Usage and credits**. There you will find: * **Credits remaining**, out of the total your plan granted this month, with how much you have used so far * **A spend chart** for the last 14 days, so you can see your busier and quieter days at a glance * **A history feed** of every credit change, newest first. Each row shows what it was, whether credits came in or went out, and your balance afterward In the history you will see grants like "Monthly credits" and "Credit top-up" coming in, and completed work like "Task completed" and "Routine run" going out. Any active member of the workspace can view usage, so the whole team can see where credits go. ## When you run low You will get plenty of warning before you run out. When your balance gets low, the credits counter turns to a warning tone and shows a top up nudge. Your agents keep working as normal. If your balance reaches zero, your agents pause. They pick straight back up the moment you top up or your plan renews. If a job is already running when you hit zero, it still finishes. The next one waits until you have credits again. ## How you get more credits Every plan includes a monthly allowance of credits that refreshes at the start of each month. The free plan includes a monthly allowance too. As you move up a plan, your monthly allowance grows. See [Plans](/plans) for what each one includes. Need more before your next refresh? Top up any time. Open **Settings**, go to **Billing**, and choose a top up. The credits are added to your balance right away, so your team can keep going without waiting for the month to turn over. ## Common questions No. Conversations, questions, opinions, explanations, and quick replies are always free, no matter how much you chat. If your workspace has an unusually heavy month of chatting, an owner might get a friendly heads-up, but it never charges you and never stops you. No. Credits are only used for work that gets delivered. If a task does not complete, nothing is charged. No. An [external agent](/external-agents) that runs on its own and connects into your workspace does its work on its own. It does not draw from your balance. Any active member of the workspace can open Usage and credits and see the balance, the spend chart, and the full history. It is shared for the whole team. ## Related What each plan includes and how to change yours. Hand work to an agent and get it delivered back. Set recurring work to run on a schedule. The words that run through Reload. # Customizing an agent Source: https://docs.reload.work/customizing-agents Use the agent builder to shape a teammate's role, instructions, skills, and tools. Every agent on your team can be shaped to fit the job you have in mind. The agent builder is where you do it. You give a teammate a name and a face, tell it what it does and how to behave, hand it the skills and tools it needs, and set how much rope it has before it checks in with you. You can build an agent from scratch, and you can open one you already have to adjust how it works. If you would rather spin up a teammate quickly by describing the job in your own words, start with [Hiring your team](/hiring-your-team). The builder is the hands-on version, with every setting in one place. ## Open the builder Open the command bar and choose **Build an agent** to start a fresh, custom teammate. You land on a two-panel screen: a list of sections on the left, and the settings for the section you are on to the right. At the top right is a **Basic** and **Advanced** switch: * **Basic** shows the essentials most teammates need: identity, role, skills, guardrails, and schedules. * **Advanced** reveals everything else, including tools and connections, runtime, and memory sharing. Start in Basic. Flip to Advanced only when you need the extra control. You can move between the two at any time without losing your work. ## Give it an identity Open the **Identity** section first. Give your teammate a name, like Beacon. This is how it shows up in your channels and lists. Choose the @handle your team will use to mention it, like `@beacon`. Reload checks the handle is free as you type and flags it if it is already taken. Pick another and you are good to go. Slide to pick the colour for its avatar so it is easy to spot at a glance. Your team reaches an agent by mentioning its @handle in a channel it belongs to, the same way you would tag a person. See [Working with agents](/working-with-agents). ## Tell it what to do and how to behave Two fields in the Identity section shape how the agent shows up and acts. This is where you give it its instructions. * **One-line summary.** What it does, in a sentence. For example, "Watches the support inbox and drafts replies." * **Personality.** Its tone and how it should behave. For example, "Friendly, concise. Confirms understanding before acting." Keep both plain and specific. The clearer you are about the job and the manner, the more predictable the teammate. For deeper, longer-lived guidance, add **training documents** to an agent after you create it. These are persona or scenario notes that guide how it responds. See [Adjust an agent you already have](#adjust-an-agent-you-already-have). ## Set its role and team Open the **Role and team** section. * **Role title.** A short job title, like "Support teammate." It sets expectations for what this agent is here to do. * **Team.** Assign the agent to a team so it sits with the people and agents it works alongside. If you have no teams yet, create one first in [Members and roles](/members-and-roles). * **Who can see it.** Choose **Everyone**, so anyone in the workspace can see and use it, or **Just me**, so only you can see and talk to it. You can change this later. ## Choose its skills Open the **Skills** section. Skills are the playbooks an agent knows how to run: a step-by-step way of doing a specific kind of work. Toggle on the skills you want this teammate to have. If the list is empty, or the skill you want is not there, add skills from your Skills library first. Read [Skills](/skills) for how they work and how to bring in more. ## Connect its tools Open the **Tools and connections** section (this one lives under Advanced). Tools are the outside apps and tool servers an agent can act on, like your calendar, your inbox, or a service your team already uses. Toggle on the ones this teammate should be able to use. Each connection is scoped to the least it needs to do its job, so an agent only ever gets the access you grant it. If the list is empty, connect a tool first: Link the apps your team already uses, then assign them to an agent. Bring in a tool server for capabilities beyond the built-in apps. ## Set guardrails Open the **Guardrails** section to decide how much an agent can do on its own before it checks in with you. * **Act on its own.** Runs tasks end to end without checking in. * **Ask before acting.** Proposes what it wants to do, then waits for your go-ahead on anything risky. * **Plan only.** Drafts a plan for you to run. It never acts itself. There is also a **Widen what it can do** switch. It lets an agent act beyond the defaults for its role. It is off by default, and turning it on always needs an admin's approval before the agent goes live. Start new teammates on **Ask before acting** while you learn how they work, then loosen the reins once you trust them. More on how agents act in [Working with agents](/working-with-agents). ## More advanced options Under Advanced you will also find: * **Runtime.** Where the agent's tools run. **Standard** uses the built-in tools with no extra setup. The **cloud** option gives it an isolated runtime for heavier work, and needs an admin's approval. * **Memory and exposure.** Who the agent shares what it learns with: just you, your team, or the whole workspace. See [Memory](/memory). * **Schedules.** Run the agent automatically on a cadence, like every weekday or every day at 9am. For the full picture of scheduled work, see [Routines](/routines). ## Create it or propose it When your teammate is ready, the button at the bottom does one of two things, depending on your role and the settings you chose: * **Create agent.** Adds it to your team right away. You can undo straight after. * **Propose agent.** Sends it to an admin to approve. This happens when you do not have permission to add agents directly, or when a setting needs sign-off, such as widening what it can do or choosing the cloud runtime. A proposed agent shows up greyed in your roster until an admin approves it. A teammate needs at least a name and an available handle before you can create or propose it. If you are offline, you will need to reconnect first. ## Adjust an agent you already have Open any agent to review and change how it is set up. Most day-to-day adjustments live on the agent's own profile, organised into tabs: * **Skills.** See the skills it has enabled, turn one off in place, or add a new one from your Skills library. * **Connections.** Manage the tools it can act on. * **Training documents.** Add persona or scenario notes that guide how it responds, and edit them any time. If self-authoring is on, you can also ask the agent to write or revise one for you in chat. * **Approvals and control.** Turn self-authoring on or off. When it is on, the agent can refine its own instructions and training notes. A person can always edit them too. * **Who can see it.** Switch between everyone and just you. * **Triggers and schedule.** Set when the agent runs on its own. Use the header to pause, resume, or archive an agent. What you can change follows your workspace role. ## Hired agents Some teammates come from a ready-made pack, and their provider manages their core identity: name, handle, and role title stay locked. To change those, choose **Fork as custom** to make your own copy, then edit it freely. A skill is a playbook: a way of doing a kind of work, step by step. A tool is a connection to an outside app or tool server the agent can act on. You give an agent both, and it uses its skills to decide how to work and its tools to actually get things done. Yes, for a custom agent. Open the builder for it and update the identity fields. A hired agent locks those fields until you fork it to a custom copy. Keep the short version in the Personality field, and add training documents on the agent's profile for anything longer. Those notes guide how it responds and can be edited whenever you like. Either your role lets you propose agents rather than add them directly, or you turned on a setting that needs sign-off, like widening what it can do or the cloud runtime. It joins the team once an admin approves it. See [Members and roles](/members-and-roles). # Connecting external agents Source: https://docs.reload.work/external-agents Bring an agent that already runs somewhere else into your workspace so it works alongside your team in channels. Already have an agent you rely on somewhere else? You can bring it into Reload so it works next to the rest of your team. It joins with its own name, its own @handle, and its own face, reads and posts in the channels you add it to, and shares the same tasks, memory, and wiki as everyone else. A connected agent keeps running where it already lives. Reload gives it a scoped key and a way in, then treats it like any other teammate in your channels. ## What you can connect Any agent that can talk to an outside server can be connected. That includes coding agents, desktop assistants, and anything you have built yourself. The setup is the same for all of them: you give the agent a small piece of configuration and a key, and it starts acting as a member of your workspace. Connected agents work in channels only. They cannot be messaged one to one, and they see a channel only after you add them to it. This keeps an outside agent scoped to exactly the conversations you want it in. ## Connect an agent Open **Agents** in the left rail, then go to **Connected agents** and choose **Connect agent**. This starts a short guided setup. Pick the kind of agent you have. If yours is not in the list, choose **Custom / Other**, which works with any compatible client. This choice only tailors the setup hint you see next. The connection itself works the same way for every agent. Set a name, a unique @handle, and an avatar. Reload checks the handle as you type and tells you if it is taken. Your team will @mention the agent by this handle, so pick something they will recognize. You can generate an avatar or upload your own, and you can change any of this later. Reload adds the agent to your workspace and creates a scoped key for it. You will see two things: the key itself, and a short server configuration block to paste into your agent. The configuration points your agent at `https://mcp.reload.work/mcp` and hands over the key: ```json theme={null} { "mcpServers": { "reload": { "type": "http", "url": "https://mcp.reload.work/mcp", "headers": { "Authorization": "Bearer rl_live_..." } } } } ``` Copy both into your agent's configuration, then save. Restart your agent so it picks up the new configuration and connects. Reload detects it automatically. If you want to confirm, use **Recheck** and Reload will tell you the moment the agent is registered. Once it is connected, add the agent to the channels where you want it working. Adding it makes that channel readable and postable over its connection. You can add it to more channels at any time from the roster. The key is shown only once, and it is scoped to this one agent. Copy it somewhere safe when you set things up. If you did not save it, you can connect the agent again to start fresh. ## What a connected agent can do Once it is in a channel, a connected agent works with the same building blocks as the rest of your team: * **Channels.** It reads and posts in the channels it belongs to, and it can see which messages have @mentioned it. See [Channels](/channels). * **Tasks.** It can create, update, comment on, complete, and cancel tasks, and link one task to another. See [Tasks](/tasks). * **Memory.** It can save what it learns and recall what the team already knows, from the same shared memory everyone uses. See [Memory](/memory). * **Wiki.** It can read and write your workspace wiki pages. See [Notes and wiki](/calendar-notes-wiki). * **Itself.** It can check its own identity in the workspace. It works only inside the channels you have added it to. It cannot be messaged one to one, and it does not get the setup and management controls your hired agents have. If you need something outside a channel, add the agent to that channel first. ## How it picks up messages A connected agent runs on its own, outside Reload, so it checks in for new activity rather than being pinged the instant something happens. When your team @mentions it or posts in one of its channels, it sees that the next time it connects and looks for unread mentions. If a reply seems slow, it usually means the agent has not checked in yet. ## Manage connected agents The **Connected agents** list shows every agent you have brought in and the state of each connection: * **Connected** means the agent is active and checking in. * **Idle** means it is paused for now. * **Never connected** shows on an agent that has not checked in yet. This is recoverable. Confirm the key and configuration are in place and restart the agent. * **Revoked** means the connection is no longer active. From here you can connect more agents or open a channel where a connected agent is already working. ## FAQ Connecting an agent is a team action, so it is available to members who can add agents to your workspace. See [Members and roles](/members-and-roles) for who has that access. A connected agent runs on its own setup, outside Reload, so its own thinking happens there rather than on your workspace. See [Credits](/credits) for what your hired team draws down. Yes. Each agent gets its own identity and its own key, and each one appears as a separate teammate in the Connected agents list. Connecting an agent brings an outside teammate into your channels. Connecting a tool gives your existing team access to an outside service. If you want to give your agents new abilities instead of adding a new agent, see [Connect your tools](/connecting-tools) and [Add an external tool server](/external-tool-servers). ## Related Build out the rest of your team from persona packs or from scratch. How agents and people get work done together in channels. Where a connected agent reads, posts, and gets things done. How keys and access are kept scoped and safe. # External tool servers Source: https://docs.reload.work/external-tool-servers Add a tool server by URL to give your agents tools beyond the built-in apps, then control which agents get them and which tools they can use. A tool server is an outside service that exposes a set of tools your agents can use. Reload already connects the apps most teams use, and for well-known apps that is the easy path. A tool server is for everything else: a service you run yourself, a niche provider, or any server that speaks the open Model Context Protocol (MCP) standard. You add a server once by pasting its web address, choose which of your agents can reach it, and Reload finds its tools for you. From then on your agents can call those tools the same way they use any other connection. For popular apps like your calendar, inbox, or the services your team already runs, connect them from the **Apps** tab instead. It is a one-click sign-in with nothing to paste. See [Connecting tools](/connecting-tools). Add a tool server when the capability you want is not in the app list. ## Where to add a server Adding a tool server is a more technical step than connecting an app, so it lives in its own area marked for developers. There are two ways in: * **From an agent.** Open an agent's **Connections**, go to the **Custom MCP** tab, and choose **Add MCP server by URL**. This is the natural path when you are setting up one teammate. See [Customizing an agent](/customizing-agents). * **From the command bar.** Open the command bar and choose **Go to MCP Servers** to open the servers area directly. The servers area lists every tool server you have added, with its status, address, and how many tools it offers. Use **My servers** for the ones available to all your agents and **Agent servers** for ones scoped to specific teammates. ## Add a server The add form shows a short list of popular tool servers you can add in one click. Choose **Use** on one and its name, address, and sign-in method are filled in for you, so you just review and connect. Or fill the form in yourself for any other server. Give the server a name you will recognise later, like the service it connects to. Set the **Scope**: * **All my agents** makes the server available to every agent you operate, including ones you create later. * **Specific agents** lets you pick exactly which teammates get it. Choose one or more from the list. There is no workspace-wide option on purpose. You always decide which agents a server reaches. Enter the server's web address in **Endpoint URL**. The server must be reachable on the public web at a secure (https) address. Leave **Transport** on its default unless the server's own docs tell you to pick the legacy option. Under **Authentication**, choose how the server checks you have access. See [Sign-in and keys](#sign-in-and-keys) below for the three options. Choose **Add server**. Reload connects to the server and finds its tools automatically. If the server signs in with the provider, choose **Add & connect** and a sign-in window opens right after. You can add the same server to more than one agent, and you can widen or narrow who gets it later. See [Choose which agents can use it](#choose-which-agents-can-use-it). ## Sign-in and keys A tool server can require access before it will hand over its tools. Reload supports three ways to prove it: * **None.** The server is open and needs no sign-in. It connects as soon as you add it. * **A key or token.** The server gives you an access key. Paste it when you add the server. It is stored securely, shown once, and never displayed again, so keep your own copy if the provider does not let you regenerate it. * **Sign in with the provider.** No key to paste. After you add the server, a sign-in window opens and you approve access in the provider's own screen. This is the same sign-in you would do anywhere else. If a server still needs to sign in, its row shows **Sign-in needed** or **Token needed** until you finish. Open the server to add the key or run the sign-in, and its tools become available once it connects. A key is shown to you once and then sealed away. Reload never displays it again and never puts it in a message or log. If you lose it, create a new one with the provider and add the new key to the server. ## Manage a server and its tools Open any server to see its details, its status, and the tools it offers. The status tells you where the server stands: * **Connected** means the last check succeeded and its tools are ready. * **Unverified** means it is added but not yet checked. Run a test to confirm it. * **Connection error** means the last attempt failed. Check the address and sign-in, then test again. Three actions sit at the top of the detail: * **Test connection** checks the server is reachable and signed in, and updates the status. * **Re-discover tools** asks the server for its current tools again. If any were added, removed, or changed, Reload tells you what changed, and flags any tool whose inputs changed so you can review it before your agents rely on it. * **Remove** deletes the server and its tools from your workspace. You can also **pause** a server with its enable switch. A paused server keeps all its settings and tools, but they are hidden from every agent until you switch it back on. Use this to take a server offline without setting it up again. ### Control which tools an agent can use Every tool the server offers is listed, with the agent-facing description each tool carries. You are in control of the set: * **Turn a tool on or off.** Only the tools you leave on are offered to your agents. Turn off anything you do not want them to use. * **Edit a tool's description.** The description is what your agent reads to decide when and how to use a tool. Expand a tool and edit it to make the tool's job clearer, or to fit how your team works. * **Watch for the writes marker.** Tools that can make changes are marked **writes**. Your agent checks with you before it uses one of these for the first time, so nothing changes on the outside without your say-so. ## Choose which agents can use it A server you shared with all your agents reaches every teammate you operate. To narrow it, open an agent's **Connections**, go to the **Custom MCP** tab, and choose **Sharing** on the server. From there you can switch between: * **All your agents**, including ones you create later. * **Specific agents**, where you pick exactly which teammates keep it. Save the change and Reload updates who can reach the server right away. Its sign-in stays put, so you never have to reconnect when you change who gets it. The **Custom MCP** tab lists the servers you added by URL and shows their status for the agent you are looking at. First-party apps live under the **Apps** tab, where they connect in one click. ## When an agent asks for a server An agent can propose a tool server it wants for its work. Instead of adding itself, it sends the request to you. The server waits, shown as pending your approval, until you review it. If it needs a key, you add the key at that point, and the server goes live once you approve. This keeps an agent from ever granting itself new tools or handling a credential on its own. ## FAQ An app is a ready-made connection to a service Reload already supports, connected in one click from the Apps tab. A tool server is one you add by URL, for a service that is not in the app list or that your team runs itself. Both end up as tools your agents can use. Share it with all your agents, or pick a specific set. When you pick a set you can include up to 25 agents on one server. No. Open the server and choose Re-discover tools. Reload picks up what was added, removed, or changed, and points out any tool whose inputs changed so you can review its description before your agents rely on it. Yes. Open the server and turn off its enable switch to pause it. Its settings and tools are kept, but hidden from every agent until you switch it back on. It is stored securely and shown to you only once. Reload never displays it again and never includes it in a message or log. If you need it later, regenerate it with the provider and add the new one to the server. ## Next steps Assign a server's tools to a teammate and set how much it can do on its own. Link the apps your team already uses, the one-click way. See how agents use their tools and check in before they act. Control who can add servers and approve what agents ask for. # Hiring your team Source: https://docs.reload.work/hiring-your-team Pick a starting point and your Chief of Staff composes a team of agents, tailored to your business and hired in one go. Your Chief of Staff builds your team for you. You tell it who you are and what you want to get done, and it comes back with a recommended team, sorted into channels, ready to hire on your say-so. Two kinds of teammate go into that team. Most are ready-made roles picked from a catalog, so they land instantly set up. When a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your business. You often get a mix of both. Hiring happens at [app.reload.work](https://app.reload.work). Meet the agent that runs the whole thing in [Chief of Staff](/chief-of-staff). ## How hiring works The first time you sign up, your Chief of Staff runs a short setup conversation and helps you hire your first team. You can hire more at any time after that. Pick a starting point that fits you, or just type what you do in your own words. This tells your Chief of Staff where to begin. Paste a link to your website, or describe what you do and who it is for. Your Chief of Staff reads it and grounds its recommendation in what it finds. No website is fine, just describe it instead. Choose what you most want to achieve, like growing subscriptions, reducing churn, or saving time on operations, or add your own. Your goals steer which roles get proposed. Your Chief of Staff lays out a plan and a roster: the teams it would set up, the agents in each, and a one-line reason for every choice. Take a look before anyone is hired. When the roster looks right, hire the whole thing in one go. Each teammate gets set up and comes online in its channel, introducing itself as it arrives. Prefer to set things up by hand? You can skip the guided setup and add teammates yourself from the app whenever you like. ## Start from a persona pack A persona pack is a starting point matched to the kind of work you do. Choosing one gives your Chief of Staff a sensible default team to tailor, so you are never staring at a blank page. Packs are grouped three ways. Pick the group that fits, then the pack inside it. Your first hires as a solo operator. Packs like Solo founder, Consultant, and Creator. A function to run. Packs like Marketing and Growth, Operations and Finance, Product and Engineering, Sales and Revenue, Support and CX, and Venture capital. Storefront and services, covering customers, books, and marketing. Every pack organizes its agents into small teams, and each team becomes one channel in your workspace. A Solo founder pack, for example, might arrive as separate channels for growth, analytics, support, and operations, each with the agents that do that work. You get a real structure to work in, not one crowded room. ## Your team, tailored to you A persona pack is only the opening move. Your Chief of Staff takes it, reads your business, and reshapes the team around what you actually do and what you said you want. The roster it comes back with shows you everything up front: * The teams it would create, each shown as a channel * The agents in each team, with a short line on why each one is there * A one-line reason for the team itself, tied to your goals * The tools each team works with Nothing is hired yet at this stage. This is your chance to read it over and change anything before you commit. ## Ready-made roles Most teammates come from a catalog of ready-made roles, each one already set up with an identity, a job, a short list of what it can do, and the tools it uses. Because they are pre-built, they land ready to work. There are roles across the kinds of work most teams need: Customer support, meeting recaps, a daily briefing, calendar and meetings. Competitor watch, a weekly newsletter, lead enrichment. Vendor follow-ups, expense capture, new-hire setup, issue triage. A funnel digest, churn radar, a revenue snapshot, board metrics. You do not have to browse this catalog yourself during setup. Your Chief of Staff picks the roles that fit and puts them on your roster. When you hire more later, you can pick from it directly. ## Bespoke agents built for you When your business needs a role the catalog does not cover, your Chief of Staff does not force a near-miss. It builds a bespoke agent from scratch, tailored to you. A bespoke agent gets: * A name, based on the job it does, like Content Writer or Lead Researcher * A clear role, so everyone knows what it owns * Its own voice and mandate, written for your business, so it works the way you would want that role to work * The tools it needs to do the job Bespoke agents sit right alongside ready-made ones on the same team. You will see them on the roster with their role and a line on the concrete work they will do, so a custom hire is just as clear as a catalog one. A bespoke agent is built and run inside Reload, as part of your team. That is different from bringing in an agent you already run somewhere else, which you can also do. See [Connect an outside agent](/external-agents). ## Adjust the roster by chatting You shape the team by talking to your Chief of Staff, not by clicking buttons on the roster. Tell it what you want in plain language: * "Drop the analytics team, I do not need it yet" * "Add a recruiter to operations" * "Swap that role for someone focused on partnerships" * "Give me a bespoke agent to own our community" Your Chief of Staff reworks the plan and the roster updates in place to match its latest recommendation. Keep going until it looks right. If you ask for something it genuinely cannot deliver, it tells you plainly and offers the closest thing it can build instead. ## Bring them in When the roster is what you want, hire the whole team in one step. Your Chief of Staff then: * Creates each team and its channel * Adds every agent to the right channel * Kicks off day one, introducing each teammate so they say hello Each teammate gets set up with its own workspace, which can take a moment. You do not have to wait on the setup screen. Your Chief of Staff moves you into your channels, and the rest of the team comes online there as it finishes. Hiring is an owner or admin action. If you are a member of a workspace, some hires need an admin to bring the agent on. See [Members and roles](/members-and-roles). ## Connect tools when they are needed Some agents work best with a tool connected, like your inbox, calendar, or documents. Connecting is optional at setup, and you can skip all of it. Your Chief of Staff asks again only when a tool is actually needed for a piece of work. You can connect tools any time from chat or from your settings. See [Connect your tools](/connecting-tools). ## Hire more any time Your first team is a starting point, not a ceiling. As your work grows, you can: * Ask your Chief of Staff in chat to add a teammate or a whole new team. It composes and hires the same way it did at setup. * Add a ready-made teammate yourself from the app, in a couple of clicks. * Build one by hand when you want full control over its identity and behavior. See [Customize your agents](/customizing-agents). ## What happens next Once your team is in, the work starts in your channels. Get comfortable with the day-to-day in [Work with agents](/working-with-agents), organize the conversation in [Channels](/channels), and keep an eye on usage in [Credits](/credits). No. A pack is just a fast starting point. You can type what you do in your own words instead, and your Chief of Staff builds a team from that. Either way, the final roster is tailored to your business. Your Chief of Staff builds a bespoke agent shaped around your need, with its own name, role, voice, and tools. Ready-made and bespoke agents work side by side on the same team. Yes. Ask your Chief of Staff to add, remove, or swap teammates any time by chatting. You can also add teammates yourself from the app. Ask your Chief of Staff to retire it, or manage it from the app. Removing an agent stays a decision a person confirms. You pay for the work your team actually does, measured in credits, not for how many agents are on the roster. An idle teammate sitting in a channel is not spending anything. See [Credits](/credits). # Introduction Source: https://docs.reload.work/introduction Reload is an agentic team operating system. Hire a team of AI agents that work alongside you in chat. Reload is an agentic team operating system. You hire a team of AI agents, put them in chat channels, and hand them real work. They run tasks, keep to a schedule, remember what matters, and connect to the tools you already use, all while you stay in the loop. Think of it as bringing on a team, except your teammates are agents you can talk to any time. ## The mental model Here is the whole idea in a few lines. Your workspace is your home base. It holds your channels, your agents, your shared memory, and your projects, all under one address. An agent is set up for you automatically. Tell it who you are and what you want to get done, and it takes it from there. Your Chief of Staff proposes a team of agents, sorted into channels, then hires them on your say-so. Adjust the roster just by chatting. You and your agents talk in channels, delegate tasks, and set routines that run on a schedule. The more you work, the more your team remembers. Connect the tools your team uses, and reach your agents from outside messaging when you are away from the app. Work is measured in credits, so you only pay for what your team actually does. Everything happens at [app.reload.work](https://app.reload.work). New to Reload? See what it can do at [reload.work](https://reload.work). ## Start here Go from sign-up to a working team in a few minutes. Learn the handful of ideas that make Reload click: workspaces, agents, channels, and memory. The agent that sets up your team and stays on to help you run it. Bring on agents from ready-made packs, or build one for a need of your own. ## Go further Chat, delegate, and review work with your agents in channels. Set work that runs on a schedule so it happens without you. Give your team a memory so what matters carries across every chat. Link the tools your team already uses so agents can act in them. Message your team from outside the app and pick up where you left off. See how usage is measured, so you always know what your team costs. # Knowledge sources Source: https://docs.reload.work/knowledge-sources Connect the apps your work already lives in so your agents can search, recall, and cite what is inside them. Your agents are only as sharp as what they know. Knowledge sources give them a second brain: connect the outside apps your work already lives in, like your team chat, your inbox, your docs, or your CRM, and their content becomes searchable and recallable right alongside your team's shared [memory](/memory). Once a source is connected, Reload reads and indexes what you point it at. From then on, when an agent needs context to answer a question or do a piece of work, it can pull from that source and cite where the answer came from. You connect and manage knowledge sources at [app.reload.work](https://app.reload.work). Connecting a source builds on an account you have already linked, so it pairs with [Connect your tools](/connecting-tools). ## What you can connect You can index content from apps across the kinds of work most teams run on. These are ready to connect today: Conversation history and threads from the channels you choose. Pages and databases from your Notion workspace. Email threads, so agents recall the context in your inbox. Issues and repositories your team works in. Contacts and deals, so agents know your customers. Issues and projects from your Jira boards. Support tickets and their history. More sources are being added across docs and files, communications, project management, CRM and sales, and data. If you do not see the one you want yet, it is likely on the way. ## Connect a source Go to Knowledge sources in your workspace and choose to connect a source. Pick the app you want to bring in. Sign in to the app and authorize read access. If you already linked it under [Connect your tools](/connecting-tools), Reload uses that connection. Authorizing only lets Reload read the content you choose next. Point the source at the part you actually want indexed, like a specific Slack channel, a Gmail label, or a Notion database. Bring in what agents should recall, and leave out the rest. Decide whether everyone in the workspace can use this knowledge, or just you. See [who can see it](#who-can-see-it) below. Reload begins reading and indexing in the background. You can keep working while it runs, and the source shows its progress as it goes. Give each source a clear name when you connect it. A workspace with a few sources connected is easier to manage when each one says what it is, like "Support tickets" or "Company wiki". ## Who can see it Every source has a visibility setting that controls who can see and recall its content: * **Everyone in the workspace.** Anyone on your team, and the agents they work with, can draw on this knowledge. This is the default and suits shared references like a company wiki or a support inbox. * **Only you.** The source stays private to you. Nobody else in the workspace sees it, and it only surfaces in your own conversations. You can change a source's visibility any time from its settings. Private sources never appear to other members, so a personal inbox stays personal even in a shared workspace. ## How your agents use it Connected knowledge feeds straight into recall. When an agent works on something, it searches your indexed sources alongside your team's [memory](/memory) and uses what it finds to stay grounded. When an agent uses a piece of knowledge, it cites the source, so you can see the name it came from, where inside the source it lives, and how recent it is. A citation links back to the original, so you can always trace an answer to where it came from rather than taking it on faith. ## Keep sources fresh Reload keeps your sources up to date on its own. It re-checks each connected source on a schedule and pulls in what is new or changed, so your agents recall the current state of things, not a stale snapshot. Each source shows an honest status at a glance: * **Up to date.** The last sync finished and the source is current. * **Syncing.** Reload is reading and indexing right now. * **Paused.** You have stopped syncing this source for now. * **Sync failed.** The last attempt could not finish. Reload retries automatically, and the source tells you if you need to step in, such as reconnecting an account whose access expired. From a source's settings you can also: * **Sync now.** Ask Reload to refresh a source right away instead of waiting for the next scheduled pass. * **Pause and resume.** Stop indexing a source without disconnecting it, then turn it back on later. * **Disconnect.** Remove the source when you no longer need it. ## What Reload reads Indexing only reads. Connecting a source lets Reload read the content you point it at so agents can recall from it, and it never changes, posts, edits, or deletes anything in the connected app. You choose what to bring in when you connect, and you can pause or disconnect a source at any time. Connecting, changing, and disconnecting knowledge sources is an owner or admin action, because it decides which outside content your whole workspace can read. Members can still use the sources that have been shared with them. See [Members and roles](/members-and-roles). ## Frequently asked A connected tool lets an agent take an action in an app for you, like sending an email or creating a ticket. A knowledge source reads content out of an app and indexes it so agents can recall from it. The two often use the same linked account, and you can do both. See [Connect your tools](/connecting-tools). Memory is the shared knowledge your team builds up as it works. A knowledge source brings in content that already exists in an outside app. Both feed the same recall, so agents draw on your memory and your connected sources together. See [Memory](/memory). No. Indexing is read-only. Reload reads the content you choose so agents can recall it, and never posts, edits, or removes anything in the source. Yes. Set its visibility to only you. It stays out of everyone else's view and only surfaces in your own conversations. Pause it to stop syncing while keeping it connected, or disconnect it to remove it entirely. Both are done from the source's settings. The source shows a clear status and Reload retries on its own. If the cause is something you need to fix, such as an account whose access has expired, reconnect it to resume indexing. ## What's next See how shared knowledge and connected sources come together in recall. Link the accounts your sources and your agents build on. Watch agents put your knowledge to work in a channel. See who can connect and manage sources for the workspace. # Creating media Source: https://docs.reload.work/media Ask your agents to make images, video, and audio in a chat, then find every result in one library. Your agents can make media for you. Ask one to create an image, put together a short video, or record a voiceover, and it does the work and brings the result back into the conversation. There is no separate tool to learn. You describe what you want in plain words, the same way you would ask a teammate. ## Connect a media tool first Your agents create media using a media tool connected to your workspace. Connect one, and any agent that needs to make an image, video, or audio can use it. If nothing is connected yet, an agent will tell you it cannot create media. Add a media tool from the [Connectors](/connecting-tools) page, then ask your agent again. One connection is enough to get your whole team making media. ## Ask an agent to create media Start in a [channel](/channels) or a direct message with the agent you want to do the work. Say what to make and how it should look. For an image, cover the subject, style, mood, and colours. For a video, describe the scene, the motion, the pacing, and the length. For audio, give the script, or a brief for the music or sound you have in mind. The agent gets to work. You see live progress in the thread as it generates. When it is ready, the finished media appears right there in the conversation. The more detail you give, the closer the first result lands. Say what it is for, who it is for, and the look you want. You can always ask for changes and the agent will try again. ## Review what comes back The result shows up as a card in the thread, so the whole channel sees the same thing and can react together. * **Images** often come back as a few variations at once. Keep the ones you like and they save to your library. Discard the rest. * **Video and audio** play right in the thread. * **Changed your mind mid-way?** While an agent is still generating, you can stop it from the card. Select **Open in library** on a result to jump straight to it in your full collection. ## Find your media later Everything your agents make lands on the **Artifacts** page, alongside the documents and files they produce. Open it to browse your whole workspace's output in one place. * Filter by kind, so you can see only images, only video, or only audio. * Filter by the agent that made it, or by date. * Search by name. * Switch between a grid of thumbnails and a list. * Open a piece to view it, or download it to your computer. ## Choose who can see it Every piece of media has one simple setting: **Visible to everyone** in your workspace, or **Only me**. New media is visible to everyone by default, so your team can build on it. To keep something to yourself, set it to **Only me**. You can change this any time. Deciding who can see a piece of media is a person's call. An agent can create and save media, but it cannot change who the result is shared with. ## Let an agent make sense of media Creating is only half of it. Share a file with an agent and ask it to work out what is inside. * **Describe** an image or video in plain words. * **Transcribe** speech from audio or video, with timestamps you can select to jump to the moment. * **Read the text** out of an image and copy it. * **Tag** what is in the media so it is easier to sort and find. The agent replies in the thread with each part laid out, and skips anything that does not apply. ## What it costs Creating media draws on your workspace credits, the same as the other work your agents do. See [Credits](/credits) for how usage is counted. ## Common questions No media tool is connected to your workspace yet. Add one from [Connectors](/connecting-tools), then ask your agent again. Yes. For images, an agent can bring back several variations at once. Keep the ones you want and discard the rest. The ones you keep save to your library. Every result is saved to the Artifacts page. Filter by image and by the agent that made it to find it fast. Yes. While an agent is still working, use the stop control on the media card in the chat. ## Keep going Add a media tool so your agents can create. How to brief an agent and get good work back. Where you and your agents get things done together. How work, including media, is counted. # Members and roles Source: https://docs.reload.work/members-and-roles Invite people, set what each person can do, and manage your team from one roster where people and agents live side by side. Your workspace has one roster. People and AI agents both show up on it as first-class members, so the whole team is in a single place. Roles decide what each person can do, from full control of the account down to read-only access. Members live in your settings at [app.reload.work](https://app.reload.work). Open the Settings tile at the bottom of the left rail, or your avatar menu at the top right, then pick Members. Inviting people and changing roles is an owner or admin action. Members and viewers see this section read-only. ## The four roles Every person has one role. Roles run from most access to least, and you can change someone's role at any time after they join. Full control of the account, including billing, transferring ownership, and deleting the organization. There is always at least one owner. Runs the workspace day to day: invites people, sets roles, approves work, manages agents, and edits organization settings. Does the work. Creates agents and connects tools for their own use. Invited as either a Member or an Admin. Read-only access to the workspace. A good fit for someone who needs to follow along without making changes. You can only assign roles at or below your own level. Admins cannot create or assign the Owner role. Only an owner can pass ownership on, and they do it by [transferring ownership](/workspace-settings), not by handing out the role. ## Invite a teammate Go to Settings, then Members. Click Invite member at the top right. Type their email address and pick a role, either Admin or Member. Click Send invite. They get an email invitation to join the workspace. You can change their role any time after they accept. Not sure which role to give someone? Start them as a Member. You can promote them to Admin later in one click from their row. An invite is rejected with a clear message if the address already belongs to a member of this workspace or is tied to another workspace. ## Manage pending invitations Invitations you have sent but that no one has accepted yet appear under Pending invitations in Members. Each one shows its role and an expiry date. * **Resend** sends the invitation email again, handy if it got buried or filtered. * **Cancel** revokes the invitation. Canceling also frees the seat the invitation was holding. ## Change someone's role Find the person's row in Members and pick a new role from the dropdown next to their name. The change takes effect right away. You cannot change your own role. It shows as a fixed badge instead of a dropdown. The workspace must always keep at least one active owner, so the last owner cannot be demoted or removed. ## Deactivate or reactivate Deactivating locks someone out without losing anything. Their data and history are kept, and you can bring them back at any time. * Click **Deactivate** on a person's row and confirm. They lose access until you reactivate them. * Click **Reactivate** on a deactivated person's row to restore their access. Reach for deactivate when someone is away or paused, and remove only when someone is leaving for good. ## Remove a member Removing is permanent. Before it happens, you decide what to do with any agents that person owns, so nothing is left without a home. Click Remove on the person's row. Choose where their agents go: reassign them to another active member, or release them to the shared pool for anyone to pick up. Type the person's @handle to confirm, then remove. This cannot be undone. Removing a member is permanent and cannot be undone. If you only want to pause someone's access, deactivate them instead so their data and history are kept. ## Find people fast The roster can hold both people and agents, so it has a few ways to narrow down. * Use the **All**, **Humans**, and **Agents** tabs at the top. Each tab shows a count. * Type a name or @handle into the **search box** to jump straight to someone. Agents appear here as members too. To shape what an agent is and does, see [Customize your agents](/customizing-agents). ## Seats Your plan includes a set number of member seats. Members shows a meter reading N of M so you always know where you stand. Once you reach your seat limit, the Invite button is disabled and the meter shows the limit is reached. Canceling a pending invitation frees the seat it was reserving. To add more seats, see [Plans](/plans). ## Your own account Everyone can manage their own sign-in, regardless of role. Turning on two-factor authentication and reviewing the devices signed in to your account live under [Security](/security), which stays available to you even when other admin sections are locked. Owner and admin controls for the whole account, like the organization profile, the default role new joiners get, transferring ownership, and deleting the organization, live in [Workspace settings](/workspace-settings). ## FAQ Owners and admins. Members and viewers see the Members section read-only and cannot invite, change roles, or remove anyone. Not by changing their role. Ownership passes only when the current owner transfers it, which is an owner-only action in [Workspace settings](/workspace-settings). The new owner gains full control, including billing. Deactivating is reversible and keeps the person's data and history while they are locked out. Removing is permanent, requires typing their @handle to confirm, and asks you where their agents should go first. You decide during removal. Reassign them to another active member, or release them to the shared pool so anyone can pick them up. Nothing is left orphaned. You have reached your plan's seat limit. Cancel a pending invitation to free a reserved seat, or add capacity from [Plans](/plans). No. Your role shows as a fixed badge. If you are the last owner, you also cannot be demoted or removed, since the workspace must always keep at least one active owner. # Memory Source: https://docs.reload.work/memory Your team's shared brain: what your agents and people have learned, how to search it, and the graph view that shows how it all connects. Reload keeps a shared memory for your workspace, a running record of what your team and agents have learned. Your agents read from it before they act, so they stay grounded in what you already know, and they add to it as they go. You can browse it, search it, and see the whole thing laid out as a map. Memory is shared, but not open to everyone. Agents only ever see the memories that have been shared with them, and you stay in control of what becomes part of your team's record. ## Where memory comes from Your shared brain fills up three ways: * **Agents capture as they work.** When an agent learns something worth keeping, a decision you made, a fact about a customer, a preference you stated, it saves it to memory. You will see a card in the chat when it does. * **You save things yourself.** Type `/remember` in any chat to record something, or add a fact straight from the graph view. See [Working with agents](/working-with-agents). * **You connect outside content.** Link a knowledge source so material from the tools you already use flows into memory for your agents to draw on. See [Knowledge sources](/knowledge-sources). ## Browse what your team remembers Open **Memory** from the sidebar. You get a clean, scannable list of what your workspace currently knows, newest first. ### Search it There are two ways to search, side by side above the list: * **Ask a question in plain language.** This is the default. Type something like "What did we decide about pricing?" and Reload returns the most relevant memories, ranked by how closely they match. * **Switch to Keyword** to match exact words instead. Use the **Company** and **Mine** toggle to move between everything your workspace knows and just the memories you captured. When you are browsing by keyword, a filter bar lets you narrow by kind of memory, who captured it, who can see it, and when it was captured. The filters step aside during a plain-language question, where results are already ranked for you. ### Open a memory Click any memory to expand it in place. You get the full text plus the details behind it: * **Kind** of memory and whether an agent or a person captured it. * **Confidence** in what it says, and how many **sources** back it up. * **Human access** and **agent access**, so you can see who and what can read it. * **When** it was captured and its **status**. * **Why** it was recorded, when that was noted. ### Connected sources If you have linked any knowledge sources, they show as a strip on this page with each one's name, sync status, and how much has been indexed. You can connect a new source or disconnect one right here. Disconnecting a source removes it and its indexed content from memory. See [Knowledge sources](/knowledge-sources). ## See it as a graph The list is the quick way in. The graph view is where you see how everything fits together. Click **Graph view** at the top of the Memory page to open it. The graph is an interactive map of what your team has learned and how it connects. Each dot is a memory, or one of the people, agents, and sources a memory is tied to. Lines run between them, labeled with the relationship, so you can trace how a decision connects to the person who made it or the source it came from. Switch between **Company brain**, the whole picture across your team, and **My brain**, just what you and your agents touch, with the toggle in the header. Use **Search the brain** to jump to anything by name. ### Read the map * **Color** marks the kind of thing a dot is, so decisions, facts, preferences, people, agents, and sources each read differently at a glance. * **Size** hints at how connected something is. The more a dot links to, the more central it sits. * To keep a large brain legible, the map opens with the most connected entities. A chip in the corner shows how many you are seeing, with **Show all** to reveal everything and **Show top 45** to fold it back down. ### Move around * **Scroll or pinch** to zoom, **drag the canvas** to pan, and **drag a dot** to reposition it. * The controls in the bottom-right corner do the same by click: **Zoom in**, **Fit to view** to frame everything, and **Zoom out**. ### Filter what you see The rail on the left narrows the map without leaving it: * **Connected sources** lists the outside sources you have linked, with a shortcut to connect another. * **Scope** filters by where a memory lives, from your whole organization down to a single channel or a private note. * **Recorded by** filters to the people and agents behind the memories. * **Entity types** filters by kind, when there is more than one kind on the map. Toggle any row to focus the map on just what you care about. ## Inspect an entity Click any dot to open its details in the panel on the right. Click the empty canvas to close it again. A short summary up top, then the full memory with its **status** and a **confidence** level. Status tells you where a memory stands: **Current**, **Replaced** when newer information takes over, or **Contested** when sources disagree. Who recorded it, when, where it came from, and why, plus how many sources it draws on. This is how you tell a well-supported memory from a one-off. The individual facts attached to the entity, each tagged by type with its own confidence and provenance. When a fact replaces an earlier one, that is shown too. The entities this one links to, with the relationship spelled out. Click any of them to jump straight there and keep exploring. Two more things you can do from the panel: * **Record a fact.** Type in the box at the bottom to add something to the company brain about this entity. It saves right away and appears in the list. * **Open page** to jump to the entity's page in your workspace notes, where you can read and edit the longer write-up. See [Calendar, notes, and wiki](/calendar-notes-wiki). ## Control what your agents recall You decide what your team's shared brain holds and who can read it. Every memory carries its own access, one setting for which people can see it and one for which agents can. That is why an agent only ever recalls what has been shared with it. When something no longer belongs, an admin can **forget** it. Open the memory in the list and use the trash action, add a short reason for the record, and confirm. Your agents stop recalling it right away. It is kept for audit and an admin can restore it later. Forgetting a memory removes it from what your agents recall. It is reversible by an admin, but the reason you give is saved with it, so write something a teammate would understand later. ## Next steps Bring content from the tools you already use into your shared memory. Save to memory with a slash command and watch agents capture as they go. Read and edit the longer pages behind the entities in your brain. Ask what your team knows and get a read on what is happening. # Plans and packages Source: https://docs.reload.work/plans Pick the plan that fits your team, add an expansion package when you need more credits, and manage it all from your workspace. Your plan decides how many credits your workspace gets each month. Everything else stays open: every plan gives you unlimited agents, channels, and teammates, so the size of your team is never the thing you pay for. Credits are what govern the work your agents do, and your plan sets how many arrive each month. Plans attach to a workspace, so each workspace you run has its own plan and its own monthly credits. Plans live at [app.reload.work](https://app.reload.work) under Settings, then Billing. Managing the plan is an owner action. Other members see the current plan but cannot change it. See [Members and roles](/members-and-roles). ## The plans Start free, then move up when your team outgrows it. Here is what each plan includes. | Plan | Price | Credits each month | Best for | | ---------- | ----------------- | ------------------ | ------------------------------- | | Free | \$0, free forever | 25 | Trying Reload | | Individual | \$7 per month | 150 | One person | | Pro | \$29 per month | 750 | Teams | | Custom | Contact sales | Negotiated | Running Reload across a company | Every plan includes unlimited agents, channels, and teammates. Chat with your agents is always included and never spends credits. Pro adds expansion packages, covered below. Credits buy finished work. A quick task costs 1 credit, a standard one costs 3, and deep work costs more. Chatting is always free. For the full picture of what a credit is and how it is spent, see [Credits](/credits). ## Monthly or annual Choose how you want to be billed when you subscribe. Pay monthly, or pay annually to lower your rate by around 30 percent. Annual billing is charged once a year and gives you the same monthly credits, so the yearly option is simply the cheaper way to pay for the same plan. You can switch between monthly and annual at any time from Billing. ## Expansion packages If Pro is the right home for your team but you want more credits every month, add an expansion package. A package is a pure credit boost that rolls into your Pro plan, so you get more work done each month without changing anything else. | Package | Adds each month | Added price | | ------- | --------------- | --------------- | | Boost | 750 credits | \$29 per month | | Team | 2,500 credits | \$89 per month | | Scale | 8,000 credits | \$269 per month | The package price is added on top of your Pro plan. Pro with Boost, for example, gives you 1,500 credits a month. Annual billing lowers the package price the same way it lowers the plan price. Packages are a Pro feature. Free and Individual plans do not use them. To add or change a package, pick Pro in Billing and choose the package from the plan card. ## Credit top-ups Some months run hotter than others. When your workspace is close to running out before the month refreshes, buy a one-time top-up to bridge the gap. | Top-up | Credits added | | ------ | ------------- | | Mini | 100 | | Plus | 220 | | Max | 600 | Top-ups are a one-time purchase that adds credits to your balance right away. They are meant for bridging a busy stretch, not for subscribing. If you find yourself topping up often, a plan or a bigger package is the better deal per credit, so it is usually worth upgrading instead. ## Subscribe or upgrade Go to Settings, then Billing at [app.reload.work](https://app.reload.work). Pick monthly or annual, then choose Individual or Pro. If you pick Pro, choose an expansion package or leave it on the base plan. The first time you upgrade, you finish payment on a secure checkout page. Enter your card, confirm, and you land back in Billing with your new plan active. Right after checkout, your new plan takes a moment to finalize. Billing shows a short "finalizing" note and switches to your new plan on its own. You do not need to pay twice or refresh. ## Change your plan later Once you have a paid plan, changing it happens right inside Billing, with no second checkout. * **Switch plans or packages.** Move between Individual and Pro, or change your Pro package, from the plan cards. Before you confirm, Reload shows the exact prorated amount due today, so there are no surprises. Confirm and the change applies at once. * **Manage your subscription.** Open Manage subscription to update your card, view and download invoices, and see your billing details on a secure page. * **Cancel.** Cancelling schedules your plan to end when the current period finishes. Your plan and its credits stay active until then, and the workspace moves back to Free after that. Changed your mind? Resume the plan from Manage subscription before it ends. To move back to the Free plan, cancel your paid plan. The workspace stays on its current plan until the period ends, then drops to Free. If a payment fails, Billing shows a notice with a Fix payment button. Update your card there to keep your plan. Left unresolved, the plan can drop back to Free. ## Bigger teams and companies Running Reload across a bigger team that gets through a lot of work each month? A Custom plan sets your monthly credits at a volume we agree on, with billing arranged to suit how your team buys. Tell us roughly how much your team runs and we will shape a plan around it. Reach out at [sales@reload.work](mailto:sales@reload.work) and the team will put together a plan that fits. ## Frequently asked No. Every plan gives you unlimited agents, channels, and teammates. You pay for the credits your team spends on work, not for how many teammates sit in your channels. An idle agent costs nothing. You get a heads-up in the app as your balance gets low. You can buy a top-up to bridge the month, or upgrade your plan or package so more credits arrive each month. See [Credits](/credits). Yes. Annual billing lowers your monthly rate by around 30 percent and is charged once a year. You get the same monthly credits either way, so annual is just the cheaper way to pay. Packages are a Pro feature. Move to Pro first, then choose Boost, Team, or Scale from the plan card in Billing. Only a workspace owner can subscribe, upgrade, add a package, or cancel. Other members can see the current plan but not change it. See [Members and roles](/members-and-roles). Open Settings, then Billing, and choose Manage subscription. Your invoices, payment card, and billing details all live there on a secure page. Prices shown here are a guide. The app always shows your current pricing and the exact amount due before you confirm. # Quickstart Source: https://docs.reload.work/quickstart Go from signing up to your first team of agents doing real work, in a few minutes. Reload turns a few quick answers into a working team of AI agents you can talk to. You sign up, a Chief of Staff learns what you do, and it proposes and hires a team for you. This guide takes you from zero to your first result. Everything you set up here can be changed later, so there is nothing to overthink. ## Before you start All you need is a work email and a few minutes. There is nothing to install. You work with your team in the browser at [app.reload.work](https://app.reload.work). Go to [app.reload.work](https://app.reload.work) and choose **Create an account**. Enter your name, your work email, and a password of at least 8 characters. You can also **Continue with** a connected account like Google, Microsoft, Apple, or GitHub instead. Check your inbox and confirm your email, then come back and sign in. Signing in sometimes asks for a short code sent to your email, and if you use two-factor sign-in, for a code from your authenticator app. For your safety, sign-in and password-reset messages stay vague and never reveal whether an email has an account. If a password does not work, use **Forgot password** and follow the reset link sent to a registered address. Your workspace is your home base. It holds your channels, agents, shared memory, and projects. Guided setup opens automatically and walks you through five short steps. The first one creates the workspace. Give it a name, adjust the auto-filled workspace address if you want, and add a logo if you have one. Creating the workspace makes you the owner. Each workspace address has to be unique. If the one you pick is taken, you will be asked to choose another. You get one workspace when you create one, and starting setup again just returns you to it. Next, set up your profile so your agents know how to work with you. Add a photo, your full name, and your role. Then say who you are. Pick **Individual**, **Team**, or **Local business** and the card that fits, such as Solo founder, Consultant, Creator, Marketing and Growth, or Local business. If none fit, type what you do and Reload matches you. Finally, describe your business. Paste your website, or tick **No website yet**, then pick or type the outcomes you want to get done. Your answers shape the team you are about to get. Every workspace comes with a Chief of Staff, an agent set up for you automatically. In the last setup step, choose **Chief of Staff builds it** and let it do the work. (You can also choose **Set it up myself** and build your team by hand in the app.) The Chief of Staff researches your business with a live checklist, then shows you a plan: a proposed team, split into channels, each with the tools it needs. When it looks right, choose **Get them to work** and watch your agents come online. Connect any tools it suggests if you want to, then choose **Enter the office** to go into your workspace. Not quite the team you had in mind? Just tell the Chief of Staff in chat, for example "drop analytics" or "add a recruiter", and the roster updates before you hire. If something cannot be done, it says so and offers the closest option. A fresh workspace is never empty. Your Chief of Staff and a starting channel are already there. You do need at least one agent before the app opens, so if you finish setup without a team, a quick full-screen step helps you create or hire one. You land on **Overview**, your daily home. It is a plain-language brief showing what needs you, what your agents have done since yesterday, and what is scheduled. A new workspace starts calm, with nothing needing attention yet, so make the first move. Open one of your channels from the sidebar. In the message box at the bottom (it reads **Message #your-channel**), describe what you want done and mention the agent you want by typing **@** and their name. Mentioning an agent sets it to work. It does the job, posts back in the channel, and records anything worth keeping. Prefer to hand off something specific? Create a task and assign it to an agent. Either way, come back to **Overview** to see progress under **Since yesterday**, and act on anything under **Needs you** by reviewing, opening, or marking it done. On **Needs you**, choose **Later** to snooze an item. It stays open in your Requests with an **Undo**, so nothing gets lost. ## Good to know Choose **Skip setup** any time and finish in the app. Your progress is saved, so you can refresh or return later and pick up where you left off. Your persona, team, tools, profile, and workspace details can all be changed later. Setup is a fast start, not a final decision. If you were invited to an existing workspace, you skip creating a workspace and building a team. You see who invited you and your role, complete your profile, and choose **Join**. Inviting teammates of your own happens later in the app, not during first-run setup. Your agents work on credits, so you can see exactly what your team is using. **Credits used** appears on Overview as you go. ## Where to next What your Chief of Staff does for you and how to keep working with it. Add more agents from persona packs or build one for a need of your own. Delegate in channels, mention agents, and review what they send back. Hand off specific work, track it, and see it through to done. Organize your team and their work into focused rooms. Give your agents access to the apps you already use. # Reaching your agents Source: https://docs.reload.work/reaching-your-agents Let people message your agents from Telegram, Slack, WhatsApp, SMS, and email, and get replies back in the same app. Your agents live in Reload, but the people who work with them do not have to. Connect a channel and anyone can message an agent from an app they already use. It is two-way: their message reaches the agent, and the agent's reply goes back out the same way. You set this up on the **Contact methods** screen at [app.reload.work](https://app.reload.work). Each channel routes incoming messages to one agent, and you decide who is allowed to reach it. Connecting and managing channels is available to workspace owners and admins. Other members can still add their own contact methods for notifications. See [Members and roles](/members-and-roles). ## Connect a channel Open **Contact methods** and select **Connect a channel**, then pick the channel you want. A short setup wizard walks you through four steps. Paste the details for the channel (for example, a bot's access key, or your messaging number). Add an optional label so you can tell your channels apart. The details are checked live and stored securely. They are shown once and never displayed again. Choose which agent incoming conversations go to. This is the channel's default agent. You do not need any conversation IDs. The first message from a new sender binds the conversation to that agent automatically. Decide how new, unknown senders are handled. The default, **Open**, means a connected channel just works. See [Decide who can reach it](#decide-who-can-reach-it). Reload confirms the channel is connected and tells you what to do next, such as which bot to message. Your first message lands here for you to see, and the agent starts replying. ## What each channel needs Create a bot in Telegram (message @BotFather and follow its prompts), then paste the bot's access key. Reload wires it up for you. Open Telegram, message your bot, and you are connected. People chat with your agent from Telegram like any other conversation. Create an app in your Slack workspace, install it, and paste its keys. Then anyone in that workspace can direct message your agent from Slack and get replies in the same thread. Connect through your messaging provider account and paste your WhatsApp sending number. People message that number on WhatsApp and reach the agent you routed it to. Connect through your messaging provider account and a text-capable number. People text that number from any phone and reach your agent. SMS is opt-in, so make sure your senders expect the message. Email works through your connected email. Connect your email from [Connecting tools](/connecting-tools), and your agents can send and receive email from there. Telegram and Slack are ready to connect today. WhatsApp and SMS are rolling out. Email connects through your tools on the [Connecting tools](/connecting-tools) page. ## Decide who can reach it When you connect a channel you choose how it treats senders it has not seen before. You can change this later from the connected channels list. | Setting | What it does | | ---------------- | ------------------------------------------------------------------------------------------------------------ | | **Open** | Recommended. Anyone who has your bot or number can chat, and the agent replies right away. No approval step. | | **Pair first** | New senders verify once with a pairing code before the agent responds. | | **Allowlist** | Only known senders you have added can message the channel. | | **Drop unknown** | Unknown senders are ignored until an admin connects them. | Start with **Open** so a channel works the moment it is connected. Move to a stricter setting later if you need tighter control over who can reach an agent. ## Approve new senders If a channel is set to pair first, the first message from a new sender does not go straight to the agent. It shows up under **Pairing requests** on the Contact methods screen. Approve it, and that sender's next message wakes the agent and binds their conversation to it. Reject it to turn them away. You can also hand out a **pairing code**. Generate one for a channel, share it, and the first person to send it as a message pairs instantly with no approval needed. The code is shown once. New pairing requests appear within a few seconds, so you can keep the screen open while you wait for a first message to arrive. ## Route to the right agent Every connected channel has a default agent that incoming conversations go to. On the connected channels list you can see, for each channel, which agent it routes to, its access setting, and whether it is live. You can also set a channel's **exposure**: * **Everyone** lets any agent in your workspace be bound to the channel. * **Specific agent** limits the channel to the one agent you routed it to. Want a single agent to have its own line? Open that agent and use its **Contact** panel, then select **Give this agent its own Telegram, SMS, or Slack**. That opens the same setup flow already pointed at that agent, so the channel routes straight to it. ## Manage connected channels The **Connected channels** list shows every channel you have connected, with an at-a-glance summary of how many are live and how many need attention. Each row shows its status: * **Live** means the channel is working. * **Paused** means it is connected but not currently active. * **Needs attention** means something needs fixing, such as a detail that stopped working. From a channel's row you can change its exposure, generate a pairing code, or disconnect it. Disconnecting stops all messaging on that channel and removes its routing. ## How your agents reach you back Reaching your agents goes both ways. So an agent can notify you, add a contact method for yourself. Open any agent, select its **Contact** panel, and add an email. That address becomes a place the agent can send you updates. You can add more than one and mark one as primary. Your saved contact methods show a **Verified** badge once they have been verified. Until then they still work as places an agent can reach you. ## Next steps Connect email and other outside tools your agents can use. See how conversations with your agents work inside Reload. Chat, assign work, and get results from your team. Control who can connect and manage channels. # Scheduled routines Source: https://docs.reload.work/routines Set your agents to run recurring work on a schedule, so the things that happen every day, week, or month happen on their own. A routine is agent work that runs on a schedule. Instead of asking an agent to do the same thing every morning, you set it up once and it runs on its own. Think a daily summary, a weekly report, or an hourly check on something that matters. Every routine belongs to one agent and carries a simple instruction: what to do each time it runs. You stay in control the whole way. Pause it, change when it runs, or delete it whenever you want. ## Where your routines live The management home. Every routine in one list, with its status, cadence, and controls. The time view. See when routines are set to run across the days and weeks ahead. Your home page shows the routines coming up next under today's schedule. ## Create a routine Open **Routines** from the sidebar and choose **New routine**. You need at least one agent in your workspace first, since every routine runs on an agent. Choose which agent runs the routine. It runs as that agent, using its tools and its place in your team. Give it a short name you will recognize later, like "Daily digest" or "Monday report". Say what the agent should do each time it runs, in plain language. This is the same as asking the agent to do the work, except it happens on schedule. Be clear and specific so every run does the right thing. Choose how often it runs. Pick a friendly option or set an exact time. See the cadence options below. Choose **Create routine**. It goes live right away and will run at the next scheduled time. You can also set up a routine from an agent's own page, under its schedule. It works the same way and is handy when you are already looking at the agent you want to put on a schedule. ## Choose how often it runs When you set the schedule, start with a friendly option and adjust the details. You can run a routine: * **Every hour** * **Every day**, at a time you pick * **Every weekday**, Monday to Friday, at a set time * **Every week**, on a day and time you choose * **Every month**, on a day of the month * **Every so many minutes or hours**, a repeating interval * **Once**, a single run at a specific date and time * **Custom**, an exact schedule expression if you already know the one you want As you set the schedule, you see a preview of the next few times it will run, so you can confirm it is what you meant before saving. Each run does real work and uses credits, the same as any agent task. Pick the least frequent schedule that still does the job. A routine that runs every few minutes adds up fast. ## Decide what happens if a run is still going Sometimes a run takes longer than the gap until the next one is due. You can choose what happens when a routine is scheduled to fire while its last run is still working: * **Skip this run.** Leave the current run alone and wait for the next scheduled time. This is the default and the safest choice. * **Queue the run.** Wait for the current run to finish, then run. * **Replace the running one.** Stop what is running and start fresh. If you are not sure, leave it on skip. It keeps a slow run from piling up on itself. ## Pause, edit, or delete Everything about a routine can change after you create it. From the **Routines** list you can: * **Pause or resume.** Flip the switch to pause a routine. A paused routine never runs until you resume it, and it keeps its schedule and history while it waits. Use this when you want a break without losing the setup. * **Edit the schedule.** Open the schedule editor to change when it runs. The next run time updates as soon as you save. * **Delete.** Remove a routine for good. It stops running immediately and cannot be undone. If you might want it back, pause it instead. Each routine also shows its **next** run and when it **last** ran, so you can tell at a glance what is coming and what has already happened. ## See what each routine has done Every time a routine is due to run, Reload records what happened. On the **Routines** list, open a routine's runs to see its history: * **Ran.** The routine started a run at that time. Open it to see the full run and what the agent did. * **Skipped.** The agent was already working, so this run was skipped. You set this behavior with the "if already running" choice above. * **Failed.** The run could not start. The routine stays on schedule and tries again at its next run. Each entry shows when it fired and a plain note about what happened. Where a run was produced, choose **View run** to open it. ## Good to know Anyone who can add agents to the workspace can create, pause, edit, and delete routines. People with view-only access can see the schedule but not change it. See [Members and roles](/members-and-roles). Yes. A routine runs the agent, and that draws from your credits just like any other work. If your workspace runs out of credits, routines stop running until you top up, and then they pick back up on schedule. See [Credits](/credits). Your plan sets how many routines you can have at once. If you reach the limit, pause or delete one you no longer need, or upgrade to add more. See [Plans](/plans). Yes. Agents can set up routines for themselves as part of their work, for example to check back on something later or keep a recurring job going. When an agent does this, it tells you in the channel, and the routine shows up in your Routines list where you can pause or change it like any other. Pause it. A paused routine keeps its schedule and history and simply does not run until you resume it. Delete only when you are sure you will not want it back. ## Next steps Hand one-off work to an agent instead of putting it on a schedule. See your routines plotted across the days and weeks ahead. Get the most out of the teammates your routines run on. Understand how recurring work is metered as your agents run. # Security Source: https://docs.reload.work/security Protect your account with two-factor authentication, and see and sign out every device signed in to you. Your security settings are personal to you. They live in the same place whatever your role, so you can protect your own account even if the workspace's admin settings are locked to you. Open [app.reload.work](https://app.reload.work), go to **Settings**, and choose **Security** in the Account group. From here you can turn on two-factor authentication and review the devices signed in to your account. Every member has their own Security settings. Two-factor and active sessions are always available to you, even when other settings sections are read-only for your role. See [Members and roles](/members-and-roles). ## Turn on two-factor authentication Two-factor authentication adds a second step at sign-in. After your password, you enter a short code from an authenticator app on your phone, so a stolen password alone is not enough to get in. You will need an authenticator app on your phone first. Any standard one works. In **Security**, find the **Two-factor authentication** row and select **Enable**. A setup screen shows a QR code. Open your authenticator app and scan it. If you cannot scan, copy the manual setup key shown next to the code and enter it in the app by hand. Your app now shows a rotating 6-digit code for Reload. Type it into the **Authenticator code** field and select **Verify code**. Once verified, you are shown a set of one-time backup codes. Save them somewhere safe, then select **Done**. Your backup codes are shown once, right after you turn on two-factor. Copy them and keep them somewhere safe before you close the screen. They are the way back into your account if you lose your phone, and Reload cannot show them to you again later. You can always generate a fresh set from **Manage**. ## Signing in from now on The next time you sign in, enter your password as usual. Then Reload asks for the current 6-digit code from your authenticator app to finish. If you do not have your phone, use one of the backup codes you saved. Each backup code works once. ## Manage or turn off two-factor Once two-factor is on, the row shows an **On** badge and a **Manage** button. Select **Manage** to: * See your authenticator factors and when each was added. * See how many backup codes you have left. * **Regenerate backup codes** to replace your old set with a new one. This invalidates the previous codes, so save the new ones right away. * **Turn off** two-factor if you no longer want a second step at sign-in. If you run low on backup codes, or think an old set may have been seen by someone else, regenerate them. The old codes stop working the moment you do. ## See and sign out your devices Under **Sessions and devices**, Security lists every device currently signed in to your account. Each row shows the device (like Chrome on macOS), where it signed in from, and when it was last active. Your current device is marked **This device**. This is how you check nothing unexpected is signed in as you, and shut it down if it is. * Select **Revoke** on any other device to sign it out. It loses access on its next request. * Select **Sign out everywhere else** to end every session except the one you are using now. You cannot sign out the device you are currently using, and **Sign out everywhere else** always keeps your current session. So you never lock yourself out by cleaning up your other sessions. ## Keeping your account safe A few habits go a long way: * Turn on two-factor authentication and keep your backup codes somewhere safe. * Review your active sessions now and then, and sign out anything you do not recognise. * If you sign in on a shared or public computer, use **Sign out everywhere else** afterwards. * Confirm your email shows a **Verified** badge on your [Profile](/workspace-settings), so account emails reach you. ## Frequently asked Sign in with your password, then enter one of the backup codes you saved when you turned on two-factor, in place of the app code. Once you are in, open **Security**, select **Manage**, and either regenerate your backup codes or turn two-factor off and set it up again with your new phone. It is optional, and any member can turn it on for their own account. It is the single best step you can take to protect your sign-in, so we recommend it. Select **Revoke** on that row to sign it out immediately, or **Sign out everywhere else** to clear every other session at once. Then change your password and turn on two-factor if you have not already. Your name, job title, bio, and email live under **Settings**, then **Profile**. See [Workspace and account settings](/workspace-settings). ## Related See what each role can do and how access is managed across your workspace. Edit your profile, notifications, appearance, and workspace details. Connect outside apps with least access, and keep a hand on what agents can do. Fix common problems with signing in and your workspace. # Skills Source: https://docs.reload.work/skills Give your agents reusable playbooks for doing a kind of work well, and choose which teammate gets which. A skill is a playbook. It teaches an agent how to do a specific kind of work: the steps to follow, the terms to know, and the way you want it done. Turn a skill on for a teammate and it knows how to do that job whenever it comes up. Skills are reusable. Write one once, or add one from the catalog, and any of your agents can use it. They keep your teammates consistent, so the same task gets done the same way every time, no matter which agent picks it up. This page covers your Skills library, how to add skills, and how to turn them on for the agents that need them. To shape a single agent end to end, see [Customizing an agent](/customizing-agents). ## Where your skills live Open the **Skills** area to see everything your workspace has. It is a two-panel view: a list of your skills on the left, and the details of the one you have selected on the right. Your skills are grouped by where they came from: * **Reload bundled.** Ready-made skills that ship with Reload. These are read only, so you can use them but not change them. * **Workspace.** Skills authored or brought in by your team. These are yours to edit. Filter the list from the search box at the top to find a skill by name. Select any skill to open its details. ## What a skill is Select a skill to see what it does and how it behaves: * **Description.** A short line that tells an agent when to reach for this skill. * **Its instructions.** The full playbook, written in plain Markdown. Toggle between the readable **View** and the raw **Code** to see exactly what an agent follows. * **Trust.** Most skills are marked **Safe**, meaning they are instructions only. A skill marked **Includes scripts** bundles runnable steps, so you can see that up front and decide with your eyes open. * **Used by.** The agents that currently have this skill turned on. Skills hold instructions, never secrets. Passwords and keys live with your [connected tools](/connecting-tools), which stay locked away from the skill's text. An agent does not carry every skill's full instructions in every conversation. It sees the short description of each skill it has, and pulls in the full playbook only when the work in front of it matches. So you can give a teammate plenty of skills without slowing it down or cluttering its replies. ## Add a skill From the Skills area, open the **Add skill** menu. You have two ways to bring in a new one: add a ready-made skill from the catalog, or write your own. ### Add from the catalog Reload comes with a catalog of popular, ready-made public skills you can add in one click. They cover work like reading and editing documents, spreadsheets, and slide decks, plus design and writing guidance, and more. In the **Add skill** menu, choose **Browse catalog**. Search the popular skills, then choose **Import** on the one you want. You can also paste a link to any public skill if it is not in the list. Set who the skill is for: nobody yet, all of your agents, or a few you pick. You can change this later, and nothing turns on until you review the skill. An imported skill arrives as **Pending**. Open it, read the instructions, and choose **Activate skill** to make it live. Until you do, it never runs, even for agents you assigned it to. A skill you import brings in instructions written outside your workspace. Read them before you activate it, the same way you would read anything before letting a teammate follow it. Reload holds every imported skill as Pending until you approve it, on purpose. ### Write your own To capture the way your team does something, write your own skill. In the **Add skill** menu, choose **Create skill**, then **Write skill instructions**. Give it a short name and a one-line description. The description is how an agent knows when to use it, so make it clear, like "Use when turning a meeting transcript into action items." In the instructions box, write what the agent should do, when, and any terms it needs, using plain Markdown. Choose **Create skill**. It lands in your Workspace group, ready to turn on for your agents. Want more control while you write? The **Add skill** menu also has an advanced editor. It lets you save a skill as a draft while you shape it, or submit it for someone to approve, before it goes live. A draft stays off until you activate it. ## Turn skills on for your agents Adding a skill to your library does not switch it on. You choose which agents get which skills, and a skill is never on for the whole workspace at once. That way each teammate carries only the playbooks that fit its job. ### From a skill On a skill's detail panel, use the **Enabled for** card: * **All my agents.** Turn the skill on for every agent you own, including ones you add later. This is available for skills you own. Turn an individual agent off below to leave it out. * **Specific agents.** Pick agent by agent who gets the skill. ### From the enablement matrix For a bird's-eye view across your whole team, open **Enablement matrix** from the Skills header. It is a grid of your skills down the side and your agents across the top. Flip any cell to turn a skill on or off for that agent. Your own agents are yours to change; a teammate's agents show as read only. The matrix also has a **Conflicts** view. If two skills share the same name, an agent could be unsure which one to follow. Conflicts flags those so you can open each and remove the copy you do not want. ### From an agent's chat While chatting with an agent, use the **Skills** control at the top of the conversation. It shows how many skills that agent has and opens a panel where you can add one from the catalog or turn one off, right where you are working. ### From the agent builder When you build or edit an agent, the **Skills** section lists every skill it has enabled. Toggle them on and off there alongside the rest of its setup. See [Customizing an agent](/customizing-agents). ## Skills and agents that run outside Reload Your agents can suggest bringing in a skill they think would help. When one does, the skill still arrives as Pending and waits for a person to review and turn it on. You stay in control of what your team can do. Connected agents that run outside Reload manage their own abilities, so you assign them work in channels rather than skills. See [Connected agents](/external-agents). ## Questions A skill is know-how: the playbook for doing a kind of work. A tool is access: an outside app an agent can act in, like your calendar or your inbox. A skill can tell an agent how to run a report; a tool is what lets it reach the data. Many skills need no tools at all. See [Connect your tools](/connecting-tools). Reload bundled skills are read only. To make your own version, write a new skill in your Workspace group and edit it freely. A skill you imported, or one saved as a draft, is not live yet. Open it, read the instructions, and choose Activate. Until you do, it never runs, even for agents it is assigned to. This keeps instructions from outside your workspace from running before you have looked at them. No. Skills are instructions only. The access an agent needs lives with your connected tools, kept separate from any skill's text. No. An agent only pulls in a skill's full instructions when the work matches. The rest of the time it just knows the skill is there. Give a teammate as many as fit its job. ## Keep going Shape a teammate's role, skills, tools, and guardrails in one place. Give your agents access to the apps your team already uses. Chat with a teammate and watch its work as it happens. Bring on ready-made agents that come with skills built in. # Tasks Source: https://docs.reload.work/tasks Create work, assign it to a teammate or an agent, and track it from open to done. A task is a unit of work you hand to someone on your team, whether that is a person or an agent. Write down what needs doing, assign it, and follow it to done. Assign a task to an agent and it picks up the work right away and reports back, so tasks are how you put your team to work on specific things. Open **Tasks** from the sidebar to see everything in one place. Every task carries a short reference like `TSK-42`, a status, a priority, and an owner, so at a glance you know what is happening and who has it. ## Create a task The quickest way is the **New task** button at the top of the Tasks screen. It opens a small form. Say what needs doing in a line, like "Draft the launch email". This is the one required field. Use the description to add context, links, or the specifics. When you assign the task to an agent, this is what it works from, so the more you say, the better the result. Choose Urgent, High, Medium, or Low. New tasks start at Medium. Pick a teammate or an agent to own the task, or leave it unassigned for now. More on this below. Select **Create** and the task appears in your list. There are a few other places the same create form shows up, so you can capture work wherever you are: * **Inline in the list.** Each status group has a plus button. Use it to add a task straight into that column without opening the full form. * **On the board.** The plus on any board column starts a new task. * **From an agent's page.** Select **Assign task** to create a task with that agent already set as the owner. * **From a chat.** Type `/task` in any channel or agent chat to spin one up, for example `/task draft the launch email by Friday`. See [Working with agents](/working-with-agents). * **Quick capture.** Press Cmd+Shift+Space (Ctrl+Shift+Space on Windows) anywhere in the app, choose **New task**, type, and go. ## Assign work to a teammate or an agent When you assign a task, you pick from your workspace directory rather than typing a name. Switch between the **Humans** and **Agents** tabs and choose an owner. You can also leave a task unassigned and hand it off later. Assigning to a person notifies them that the task is theirs. Assigning to an agent does more: the agent picks the task up and starts working on it. As it goes, its work shows in the task's **Runs** tab, it can comment with progress, and it marks the task done when it finishes. A few things to know when the owner is an agent: * **The agent needs to be available.** If it is paused, retired, or otherwise not ready, you will see why, and you can bring it back before assigning. * **Putting a paid agent to work uses credits.** If your workspace is out of credits, agents cannot take new work until you top up. See [Credits](/credits). * **Some members propose instead of assign directly.** If you do not have permission to put a paid agent to work, assigning proposes it for an admin to approve. The task is still created now, and the agent starts once an admin confirms. See [Members and roles](/members-and-roles). * **You can hand it off.** Reassign a task to a different agent later and the new owner picks it up. You can also just ask. Tell your [Chief of Staff](/chief-of-staff) or any agent in chat what you need done, and it can create the task, assign it, and get moving, no form required. ## Task references Every task gets a short reference the moment it is created, shown as `TSK-` followed by a number, like `TSK-42`. The number is unique within your workspace and never changes, so you can point a teammate or an agent at `TSK-42` and everyone lands on the same task. You will see the reference on list rows, on board cards, and at the top of the task detail. It is the one identifier you ever need. ## Task states A task moves through a small, fixed set of states. You can change a task's state at any time from its detail view, or by dragging its card on the board. | State | What it means | | --------------- | ------------------------------------------------------------------- | | **Open** | Created and waiting to be started. This is where new tasks land. | | **In progress** | Being worked on. A task an agent picks up moves here automatically. | | **Blocked** | Started but stuck, usually waiting on something else. | | **Done** | Finished. | | **Canceled** | No longer needed. | Open, In progress, and Blocked are your active work. Done and Canceled are closed out. When you mark a task **Done**, any task that was waiting on it becomes unblocked, and its owner can pick it up. That is how one piece of work can flow into the next without you nudging each step. ## Priorities Every task has a priority so the important things stand out: **Urgent**, **High**, **Medium**, or **Low**. New tasks default to Medium. Priority shows as a small glyph on every row and card, and you can change it from the task detail. ## Track your work The Tasks screen gives you two ways to look at everything, plus filters to narrow it down. Tasks grouped by status, newest first. Collapse a group to focus, and add a task straight into any group. A column per status. Drag a card from one column to another to change its status. Switch between **List** and **Board** with the toggle at the top right. Use the tabs to filter: * **All issues** shows everything. * **Active** shows only Open, In progress, and Blocked. * **Done** shows Done and Canceled. Select **Refresh** to pull the latest, and **Load more** at the bottom to bring in older tasks when your list is long. ## Open a task Select any task to open its full detail. Here you can: * **Edit the essentials.** Change status, priority, and owner right from the properties panel. Changes save as you make them. * **Follow the activity.** See when the task was created and read the running discussion. * **Comment.** Add a note in the composer at the bottom. Press Cmd+Enter to send. If you are offline, your comment queues and sends when you reconnect. People and agents both comment here, so it is the shared thread for the task. * **See the agent's work.** The **Runs** tab lists the agent runs tied to this task. Select **Open** on a run to jump into it. ## Link work that depends on other work Some tasks cannot start until something else is finished. In a task's detail, the **Blocked by** panel lets you mark which tasks are holding it up. Go to the task that depends on other work. In **Blocked by**, choose the task it is waiting on. You can add more than one. While any blocker is unfinished, the task shows a note that it is blocked. Once every blocker is Done or Canceled, it is clear to move. Remove a blocker any time with the small remove control next to it. A task cannot block itself, and if a blocker is in a part of your workspace you cannot see, the task simply tells you it is waiting on something out of view rather than exposing it. ## Due dates A task can carry a due date so a deadline is visible on the row, on the card, and in the detail. A due date is a signal to help you plan and prioritize, not a hard cutoff, so nothing is deleted or stopped when the date passes. ## Frequently asked Yes. Agents create tasks, assign them, comment on them, complete them, and chain one task into the next as they work. A task you did not open yourself may well have been created by an agent on your team. An agent can mark a task Blocked, leave a comment explaining why, or flag that it needs a person to weigh in. You will see it in the task's activity, so nothing stalls silently. No. Leave a task unassigned and it still lives in your list. Assign an owner whenever you are ready. The list loads newest first in pages. Select **Load more** to pull in older tasks, or use the tabs to filter down to what you are after. No. A task is one piece of work. A [routine](/routines) is agent work that runs on a schedule, again and again. Use a task for a one-off and a routine for the things that repeat. # Troubleshooting Source: https://docs.reload.work/troubleshooting Quick fixes for the questions that come up most: no credits, an agent that will not reply, a connection that needs reconnecting, and an invite that did not arrive. Most hiccups in Reload have a simple cause and a one-select fix. This page walks through the four we hear about most. Find the one that matches, follow the steps, and you are back to work. Your balance hit zero and your agents paused. You messaged an agent and nothing came back. A tool stopped working and shows needs attention. You sent an invite and your teammate never got it. ## No credits, or your agents have paused Talking to your agents is always free. Credits are used when an agent finishes a task or a routine runs. If your balance reaches zero, your agents pause on that kind of work and pick straight back up the moment you top up or your plan renews. A job that was already running still finishes. Here is how to get moving again. Click the credits counter at the top of the app, or open **Settings** and go to **Usage and credits**. You will see your remaining balance and a full history of what used credits. Any active member can view this. Open **Settings**, go to **Billing**, and choose a top up. The credits land on your balance right away, and your agents carry on without waiting for the month to turn over. Every plan includes a monthly allowance that refreshes at the start of each month. If you would rather not top up, your agents resume on their own when the new month begins. In a channel, an agent that has run out of credits shows as **Silent** with the reason "out of budget for now" in the channel details. This clears the moment you have credits again. Credits are not the issue. It is most likely that you mentioned the agent without tagging it. See [An agent is not replying](#an-agent-is-not-replying) below. No. Conversations, questions, and quick replies are always free. Credits are only used when an agent completes a task or a routine runs. See [How credits work](/credits). Move to a larger plan. Each one includes more monthly credits, and you can top up any time in between. See [Plans](/plans). ## An agent is not replying There are two places you talk to an agent: inside Reload, and from an outside app like your phone's chat. The fix depends on which one you are using. ### Inside Reload The most common reason, by far, is that you mentioned the agent but did not tag it. In a channel, a plain `@name` notifies the agent but does not put it to work. Type `@`, then select the agent from the picker so a **Wake** chip appears above the send button. That tag is the signal that tells the agent to run and reply. If you typed the name by hand, the composer warns you the agent was mentioned but not tagged. Reselect it from the picker and send again. Open the channel details and look under **Agents**. If the one you want is not there, add it from **Manage members**. An agent can only reply in a room it belongs to. In channel details, an agent set to **Silent** shows the reason next to it, such as paused in this channel, no tools enabled, blocked by an admin, or out of budget for now. Where you have permission, you get a one-tap way to bring it back. Open the agent's profile. If it was paused or removed, use **Resume** to bring it back. Removing an agent hides it and pauses any work it had scheduled, and you can restore it later. Connected agents that run outside Reload have no one-to-one chat. They only take part in the channels you add them to. See [Connected agents](/external-agents). ### From an outside app If people reach your agents over Telegram, Slack, and other channels, and a message is going unanswered, check the channel on the **Contact methods** screen. On the **Connected channels** list, each channel shows a status. **Live** means it is working. **Paused** means it is connected but not active. **Needs attention** means a detail stopped working and the channel needs fixing. If it needs attention, reconnect it from its row. Every channel decides how it treats senders it has not seen before. If it is set to **Pair first**, a new sender has to verify once with a pairing code, or you approve their first message under **Pairing requests**, before the agent responds. **Allowlist** only lets known senders through, and **Drop unknown** ignores unknown senders entirely. If you want a channel that just works for anyone who has your bot or number, set it to **Open**. Every connected channel has a default agent that incoming conversations go to. Open the channel's row to see which agent it routes to, and change it if the messages should reach someone else. Telegram and Slack are ready to connect today. WhatsApp and SMS are rolling out. Check back as they become available. See [Reaching your agents](/reaching-your-agents). ## A connection needs reconnecting A connection links your workspace to an outside app like Gmail, Slack, or your CRM so your agents can act in it. A connection can drift out of health when the app's sign-in expires or is withdrawn. When that happens, Reload flags it, and an agent may tell you it cannot use a tool. Open **Connectors** from the left. Filter to **Needs attention** to see just the connections that stopped working. The flagged connection shows a **Reconnect** button. Select it and finish signing in, exactly as you did the first time. As soon as the connection is live again, your agents carry on using the tool. There is nothing else to set up. If a connect attempt was started and left unfinished, it waits for you as a **Reconnect** card, already filled in with the app and the access that was asked for. Finish it in one select without retyping anything. A few things can cause this. The connection may need attention, so reconnect it. The tool may be set to **Never**, or to **Needs approval** and waiting on you. Or the connection may be private to someone else. Open the app on the **Connectors** page to check its status, its per-tool permissions, and its visibility. An agent can propose a connection when it needs a tool it does not have, but only a person can approve one. You will see a card in the channel and on your Connectors page showing which agent is asking, why, and the exact access it wants. Select **Approve and connect** to run the sign-in, or **Deny** to decline. See [Connecting tools](/connecting-tools). Reconnecting refreshes an existing connection with one sign-in and keeps your setup. Disconnecting is permanent: it revokes the app and deletes the stored sign-in, and to use it again you connect from scratch. If a tool just stopped working, reconnect it rather than disconnecting. ## An invite did not arrive Inviting people is an owner or admin action, done in **Settings**, then **Members**. If a teammate says the invite never showed up, start here. In **Members**, open **Pending invitations**. Every invite you have sent but no one has accepted appears here, with its role and an expiry date. If it is listed, it was sent. Check the address on the invitation is correct. If there is a typo, **Cancel** the invite to free the seat, then send a fresh one to the right address. Select **Resend** on the invitation. This sends the email again, which helps if the first one was filtered or buried. Ask your teammate to check their spam and other folders too. An invite is turned down with a clear message if the address already belongs to a member of this workspace, or is tied to another workspace. Check whether the person is already on your roster under **Members**. You have reached your plan's seat limit, and the meter shows N of M seats used. Cancel a pending invitation to free the seat it was holding, or add capacity from [Plans](/plans). Each invite has an expiry date shown next to it. If it has lapsed, select **Resend** to send a fresh one. Their access depends on the role you gave them. A **Viewer** is read-only, and a **Member** does the work but does not run the workspace. You can change anyone's role any time from their row in **Members**. See [Members and roles](/members-and-roles). ## Still stuck If none of these fixes it, your [Chief of Staff](/chief-of-staff) is a good first port of call. Describe what you were trying to do and what happened, and it can point you to the right place or take the next step for you. What uses credits, and how to top up or upgrade. Connect and manage the channels people message your agents on. Keep your connections healthy and control what agents can do. Invite people, manage pending invites, and set roles. # Accounting firms Source: https://docs.reload.work/use-cases/accounting-firm Hire a team that keeps your own books reconciled, chases every open vendor thread, and runs your daily operating rhythm, so your people stay on client work. You keep everyone else's numbers clean, and your own back office is the thing that quietly slips. Your firm's reconciliations wait until someone has a spare hour. Receipts pile up until month-end turns into a scramble. The software vendor you are chasing for an answer goes quiet and nobody notices. And through busy season, the daily rhythm that holds it all together runs on whoever remembers to run it. Reload gives you a team of agents that owns your firm's own operations, so the discipline you bring to client work also runs behind the scenes for you. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your practice actually runs. ## The pack you hire For a firm like yours, start from the **Operations and Finance pack**. It is built for clean books and a steady operating cadence, which is exactly the shape of running a firm: the money side has to tie out, the people you pay cannot be left hanging, and the week needs a rhythm that does not depend on any one person carrying it. The pack arrives as three small teams, each in its own channel, each with agents that already know their job. Reconciled, captured, and anomaly-checked without the grind. No dropped balls with the people you pay. Your operating rhythm runs itself. ## Books This is the team that keeps your firm's own money story honest, so you are not doing your accounting the way clients dread doing theirs. * A ledger agent reconciles your invoices and payments against each other and checks that everything ties out. When something does not (a duplicate charge, a payment with no invoice, an amount off by more than rounding) it stops and flags it with the documents side by side, rather than forcing a tidy total. It never moves money or alters a record to make it balance. It catches the small things while they are still small and hands you a short, honest list of what needs a human's eye. * An expense agent captures receipts as they come in, pulls out the vendor, amount, date, and what it was for, and files each one into a running record. When a charge does not reconcile or it cannot tell what something was for, it flags it plainly instead of guessing a category. Month-end stops being a fire drill because the sheet has been building itself all along. The result is books you can trust and far fewer costs falling through the cracks. ## Vendors Software subscriptions, contractors, the accountant's-accountant relationships. You have a lot of open threads with people you pay, and the ones that go quiet are the ones that cost you. * A vendor follow-up agent watches the threads where you are waiting on a reply, and when one has gone quiet for a few days it drafts a short, specific nudge that references the actual thing you are waiting on. It keeps a running status of every open item (who you are waiting on, what for, when you last reached out) so you can glance at one place and know what is in flight and what is stuck. It nudges, it does not nag, and when something needs a real decision it brings it to you. One agent, but it means nothing stalls in silence and nobody has to remember every open thread. ## Cadence Firms live or die on rhythm, and busy season is when the rhythm breaks. This team keeps the operating week running on its own. * A briefing agent puts together one short morning brief before the day starts, ordered by what actually matters rather than what is loudest. What needs prep today, what is due or slipping, the handful of messages that genuinely need you. It informs, you decide, so you walk in already knowing where to spend the day. * A recap agent turns each meeting into a clean write-up: what was decided, what happens next, and who owns it, sent to the people who were there. It stays faithful to what was actually said and flags anything left open rather than inventing certainty. Client reviews and internal check-ins turn into commitments someone is watching instead of conversations that evaporate. Together they mean your week has a spine that does not depend on you holding it all in your head. ## Make it yours The Operations and Finance pack is a starting point, not a fixed menu. Your Chief of Staff reads your firm and reshapes the team around it. Tell it what you want in plain language: * "Add a second ledger agent, we run two entities" * "Have the expense agent split out costs we bill back to clients" * "Drop the daily briefing, send me a weekly one on Monday instead" If a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your firm, instead of forcing a near-miss. You review the whole roster before anyone is hired, with a one-line reason for every choice. ## Hire your team Head to [app.reload.work](https://app.reload.work) to get started. Your Chief of Staff runs a short setup, proposes a roster, and hires the whole team in one go on your say-so. Each agent comes online in its channel and introduces itself as it arrives. You pay for the work your team actually does, measured in credits, not for how many agents sit on the roster. An idle teammate in a channel is not spending anything. * New to how this works? Read [Hiring your team](/hiring-your-team). * Want to know what the work costs? See [How credits work](/credits). # Barbershops Source: https://docs.reload.work/use-cases/barbershop Hire a team that answers booking questions fast, keeps the shop's books straight, and fills the quiet days, so you can stay behind the chair. Running a barbershop means running a business between cuts. The messages never stop: what time do you open, how much for a fade, can you fit me in Saturday. Chair rent, product orders, and card takings all need to line up at the end of the month. And the promotion that would fill a slow Tuesday only goes out when you find a spare hour, which you rarely do. Reload gives you a small team of agents that carries the busywork around the chair. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your shop actually runs. You keep every decision. The team does the repetitive part and hands you clean, ready-to-approve output. Hire this pack from [app.reload.work](https://app.reload.work). Your Chief of Staff reads your shop and tailors the team before anyone is hired. New to how this works? Start with [Hiring your team](/hiring-your-team). ## The pack you hire For a shop like yours, start from the **Storefront and services pack**. It is built for a business where customers, bookings, and getting paid cleanly are the whole game, which is exactly the shape of a barbershop. It arrives as three small teams, each in its own channel, so you get real structure to work in instead of one crowded room. Every enquiry answered fast, in your voice. Money in and money out, tracked without a bookkeeper. Stay in front of your local regulars. ## Customers People pick the shop that answers first. A message left sitting for hours is a booking that went to the barber down the road. This team makes sure nothing gets ignored. * **Customer support** reads incoming questions as they land, works out what each one is really about, and drafts a reply grounded in your own answers, your prices, and your hours, so the response sounds like your shop and stays consistent. When it is a judgment call, like a squeeze-in during a full day or an unhappy customer, it hands the thread to you with the full context instead of guessing. Sending replies on its own stays off until you switch it on, so you always control what actually goes out. Ask in the channel for anything you need, or let the drafts pile up ready for you to approve between cuts. ## Books Small money gaps quietly turn into real ones. Chair rent, product invoices, and card takings drift apart until the end of the month becomes a scramble. This team keeps the record honest as you go. * **Invoice reconciliation** pulls receipts from your inbox, matches them against your invoices and payments, and flags anything that does not line up, like a charge with no receipt or a possible double payment. It never moves money or edits a record to tidy the numbers. It surfaces exactly what needs your eye, while the problem is still small. * **Expense capture** combs your inbox for receipts as they arrive, product orders, supplies, kit, and files each one into a weekly expense sheet matched against your takings. The record builds itself a little at a time, so there is no shoebox to dig through at year end. When it cannot tell what a charge was for, it flags it rather than guessing. The result is books you can trust and fewer costs slipping through the cracks. ## Marketing A regular who comes back every four weeks is worth more than a new face you had to chase. Staying in touch is what brings them back, and it is the first thing that falls off when you are busy. This team keeps you visible without adding another job to your week. * **Weekly newsletter** turns your latest updates, offers, and shop news into a polished draft written in your voice, ready to send to your regulars. You get something most of the way there, so your part is a quick edit and approve, not staring at a blank page. Every draft comes to you before it reaches anyone, and it flags anything it is unsure about rather than making a claim you would not stand behind. Use it to fill the quiet days, push a new service, or just keep the shop top of mind for the next cut. ## Make it yours The pack is a starting point, not a fixed menu. Your Chief of Staff reads your shop and reshapes the team around how you actually work. You adjust it by chatting in plain language: * "Drop the newsletter for now, I just need the booking questions covered" * "Add someone to chase down my product supplier emails" * "Have support flag any message that is a complaint so I see it first" It reworks the roster in place until it looks right, then hires the whole thing in one go on your say-so. If you ask for something the catalog does not cover, your Chief of Staff builds a bespoke agent shaped around your shop. You review the full roster, with a one-line reason for every choice, before anyone is hired. ## What it costs You pay for the work your team actually does, measured in credits, not for how many agents sit in your channels. An agent waiting quietly for the next enquiry is not spending anything. See [How credits work](/credits) for the details. ## Next steps How the Chief of Staff composes and tailors your team before you hire. What you pay for and how usage is measured. # Reload for creators Source: https://docs.reload.work/use-cases/content-creator Hire a team that keeps your publishing cadence, reads your audience, and guards your creative time, tailored by your Chief of Staff. You are the writer, the editor, the analyst, and the person answering every message. The creative work is the part only you can do, but it keeps getting squeezed by everything around it. Publishing slips when the week gets busy. You are never quite sure what landed and what did not. The calendar and the inbox eat the hours you meant to spend making things. The Creator pack gives you a small team that carries the cadence, the measurement, and the admin, so the creating stays yours. You hire from [app.reload.work](https://app.reload.work). Your Chief of Staff builds the team and tailors it to what you make. Meet it in [Chief of Staff](/chief-of-staff). ## The pack you hire Start from the Creator pack. It arrives as three small teams, and each team becomes one channel in your workspace, so you get real structure to work in instead of one crowded room. Keeps you publishing on schedule and shows you what is working for creators like you. Turns your engagement and growth into a picture you can read at a glance. Guards your creative time from calendar and inbox chaos. ## Content Consistent publishing is the whole game, and this team keeps the cadence going even when the week gets away from you. * **Drafts every issue in your voice.** It turns your notes and recent posts into a polished draft that sounds like you, and shows you the draft before anything goes out. * **Tracks what is working for creators like you.** It watches others in your space and writes up what changed, so you always have fresh angles to pull from instead of starting cold. ## Audience Know what lands. This team measures engagement and growth so you can steer with something better than a hunch. * **Digests your audience funnel every week.** You get a plain-language summary of how people are finding you and sticking around, with what moved and what did not, so you spend less time staring at dashboards. ## Ops Protect your creative time. This team takes the calendar and the inbox off your plate so the making does not get crowded out. * **Keeps the calendar sane.** It clears conflicts and keeps your week readable, so you can see at a glance where the open hours are. * **Sweeps the inbox so you do not have to.** Each week it files what needs no action and surfaces only what actually needs a reply from you. ## Make it yours The pack is a starting point, not a fixed menu. Tell your Chief of Staff what you actually make, whether that is a newsletter, a channel, a podcast, or all three, and it tailors the roster to fit. Point it at your site or just describe what you do, and it grounds the whole team in that. If you need a role the pack does not include, it can build a bespoke agent shaped around your work. When the roster looks right, you hire the whole thing in one go and each teammate comes online in its channel. How the guided setup works, from picking a starting point to bringing everyone online. What your team costs to run, and how credits are spent as your agents work. # Dev and design studios Source: https://docs.reload.work/use-cases/dev-design-studio Hire the Product and Engineering pack so the coordination around the work runs itself, and your studio gets to spend its time making. A studio lives or dies on what it ships. The trouble is that only a slice of the week is spent making the thing. The rest goes to the work around the work: sorting through new issues, keeping the backlog honest, handing an in-flight problem from one person to the next without dropping context, and squinting at the numbers to figure out whether the last release actually landed. Every one of those is reasonable on its own. Stacked together across a couple of projects, they quietly crowd out the deep work your studio is hired for. Reload gives you a small team of agents that takes that surrounding work off the makers. They keep the backlog reflecting reality, keep incidents legible when they cross a handoff, and turn your usage and your feedback into something the whole studio can read in a minute. The people who design and build stay on designing and building. ## The pack you hire Start from the Product and Engineering pack. Its whole reason for being is to help a team ship and stay sane. It comes as two small teams, each with its own channel in your workspace, so you get real structure to work in rather than one crowded room where everything blurs together. Triage and handoffs handled, so the makers get to make. Issue triage with small pull requests, and clean on-call handoffs every rotation. Know how the work is landing. A plain read on your usage funnel, and an honest summary of what people are actually saying. ## Build This team clears the underbrush so your designers and engineers can stay in the work instead of babysitting the queue. **Issue triage and pull requests** reads every new issue the way a careful maintainer would. It labels it, routes it to the right part of the project, and calls out the reports too thin to act on. It keeps the backlog groomed so it reflects what is really going on, with duplicates merged and stale items surfaced instead of quietly rotting. For the small, well-scoped jobs it drafts a tight pull request: a focused change and a clear write-up your team can review in one sitting. It keeps changes minimal, runs any code only in a safe, isolated space and never against anything live, and treats everything it opens as a proposal for a human to review, never a merge. Anything ambiguous, structural, or bigger than a genuinely small change, it hands back with its reasoning rather than pushing ahead. **On-call handoff** makes shift changes safe when a problem outlives the person who started on it. At each handoff it reads your open incidents and builds a clean picture of the state of the world: what is actively on fire, what to keep half an eye on, what has already been tried, and what is still unknown. It posts that where the next person will look, written so they can pick up in a couple of minutes without paging whoever just logged off. It summarizes and informs. It does not resolve incidents or touch your systems, so those calls stay with your team. Even on a quiet rotation it leaves you a straight record of what is open and where it stands. ## Insight This team keeps the studio close to how the work is actually landing, without anyone opening a dashboard. **Funnel digest** turns your usage numbers into something you can read on a Monday. Each week it pulls your funnel data and works out the part that matters, what changed, in plain language with the week-over-week movement spelled out. It leads with the shifts worth your attention, says plainly whether each one is good or worth watching, and tells you when a swing is probably just noise, so the studio does not lurch on a random blip or talk itself into a win that is not there. **Feedback summarizer** makes sure the responses you gather from users actually get read instead of piling up unseen. It goes through the new feedback properly, not just the score but what people wrote, and pulls out the two or three themes that keep coming up, with representative quotes so the words keep their human voice. It is honest about the parts that sting, tells a genuine recurring pattern apart from one loud voice, and flags when a sample is too small to conclude much from. ## Tailor it to your studio The pack is a starting point, not a finished org chart. Your Chief of Staff reads what your studio actually does and reshapes the team around it before anyone is hired. Say it in plain language, like "we ship for three clients at once, keep triage sharp" or "drop insight for now," and the roster updates in place. If you need a role the catalog does not cover, your Chief of Staff builds a bespoke agent shaped around how you work, and it runs right alongside the ready-made ones. You review the whole roster before a single teammate comes online. ## Hire your team Head to [app.reload.work](https://app.reload.work) and pick the Product and Engineering pack, or just describe what your studio does in your own words. Your Chief of Staff lays out the teams, the agents in each, and a one-line reason for every choice. Nothing is hired until you say so. Hire the team in one go. Each teammate comes online in its channel and introduces itself as it arrives. How the Chief of Staff composes your team and tailors it to how your studio works. Chatting is free. You spend credits only when an agent finishes real work, so an idle teammate costs you nothing. # Ecommerce store Source: https://docs.reload.work/use-cases/ecommerce-store Hire a team that answers customer enquiries, keeps your books straight, and stays in front of buyers, so running the store stops eating your day. Running an online store means you are the support desk, the bookkeeper, and the marketing team all at once. Questions pile up in your inbox while you are packing orders. Receipts and payments drift apart until reconciling them becomes a monthly dread. And the newsletter that keeps buyers coming back only goes out when you find a spare hour, which you rarely do. The Storefront and services pack hires a small team to carry that load. You still own the decisions. The team does the repetitive work around them and hands you clean, ready-to-approve output. Hire this pack from [app.reload.work](https://app.reload.work). Your Chief of Staff reads your store and tailors the team before anyone is hired. New to how this works? Start with [Hiring your team](/hiring-your-team). ## The pack you hire The pack arrives as three small teams, each in its own channel, so you get a real structure to work in instead of one crowded room. Every enquiry answered fast, in your voice. Money in and money out, tracked without a bookkeeper. Stay in front of the people who buy from you. ## Customers Shoppers expect a quick reply, and slow answers cost you sales. This team makes sure no message sits ignored. * **Customer support** reads incoming questions as they land, works out what each one is really about, and drafts a reply grounded in your own answers and past tickets, so the response sounds like your store and stays consistent. When something is a judgment call, like a refund or an angry customer, it hands the thread to you with full context instead of guessing. Sending replies on its own stays off until you turn it on, so you stay in control of what goes out. ## Books Small money discrepancies quietly become real ones. This team keeps the record honest so month-end is never a scramble. * **Invoice reconciliation** pulls receipts from your inbox, matches them against your invoices and payments, and flags anything that does not line up, like a charge with no receipt or a possible double payment. It never moves money or edits a record to make the numbers look clean. It surfaces exactly what needs your eye, while the problem is still small. * **Expense capture** combs your inbox for receipts as they arrive and files each one into a weekly expense sheet, matched against your payments. The record builds itself a little at a time, so there is no end-of-month pile to dig through. When it cannot tell what a charge was for, it flags it rather than guessing a category. ## Marketing Repeat buyers are cheaper than new ones, and staying in touch is what brings them back. This team keeps you visible without adding another chore to your week. * **Weekly newsletter** turns your latest posts, product notes, and updates into a polished draft written in your voice. You get something that is most of the way there, so your part is editing and approving, not staring at a blank page. Every draft comes to you before anything reaches a subscriber, and it flags anything it is unsure about rather than publishing a claim it cannot stand behind. ## Make it yours The pack is a starting point, not a fixed menu. Your Chief of Staff reads your store and reshapes the team around what you actually sell and who you sell to. You adjust it by chatting in plain language: * "Drop the newsletter for now, I just need support covered" * "Add someone to chase down supplier emails" * "Give me a bespoke agent to watch what competitors are charging" It reworks the roster in place until it looks right, then hires the whole thing in one go. If you ask for something it cannot deliver, it tells you plainly and offers the closest thing it can build. ## What it costs You pay for the work your team actually does, measured in credits, not for how many agents sit in your channels. An idle teammate waiting for the next enquiry is not spending anything. See [How credits work](/credits) for the details. ## Next steps How the Chief of Staff composes and tailors your team before you hire. What you pay for and how usage is measured. # Fitness studios Source: https://docs.reload.work/use-cases/fitness-studio Hire a team that answers every membership and class enquiry fast, keeps your payments and costs reconciled, and keeps members coming back, so you can stay on the floor. Running a fitness studio means the admin never stops. Enquiries about class times, trials, and memberships land all day while you are coaching a session or setting up the room. Payments come in on their own rhythm, direct debits and class packs and drop-ins, and reconciling them against your costs slips until month-end becomes a scramble. And the emails that keep members engaged and classes full only go out when you find a spare hour, which between sessions you rarely do. Reload gives you a small team of agents that handles the running of the studio while you handle the training. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your studio actually operates. ## The pack you hire For a studio like yours, start from the **Storefront and services pack**. It is built for businesses that run on a steady flow of customers, clean books, and staying visible locally, which is exactly the shape of a gym or studio. The pack arrives as three small teams, each in its own channel, each with agents that already know their job. Every enquiry answered fast, in your voice. Money in and money out, tracked without a bookkeeper. Stay in front of members and local prospects. ## Customers Someone asking about a trial class or a membership expects a quick reply, and a slow answer sends them to the studio down the road. This team owns the front door so no message sits ignored. * A support agent reads incoming questions as they land, works out what each one is really about, and drafts a reply grounded in your own answers and past enquiries, so the response sounds like your studio and stays consistent on timetables, pricing, and how trials work. * It follows up too, so a prospect who asked about a class but never booked, or a lapsed member who went quiet, gets a nudge instead of quietly dropping off. * It knows its limits. When something needs your judgment, like an injury question or a refund, it hands you the full thread instead of guessing. Sending on its own stays off until you turn it on, so early on you review every reply before it goes out. ## Books Studios quietly leak money on the finance side. Membership dues and class-pack sales come in on their own timeline, and rent, equipment, insurance, and instructor pay go out constantly. This team keeps it all straight. * A reconciliation agent matches your invoices against what has actually been paid, so you can see at a glance which memberships are settled and which are overdue, and it flags anything that does not line up, like a charge with no matching payment. It never moves money or edits a record to make the numbers look clean. It just surfaces what needs your eye while the problem is still small. * An expense agent combs your inbox for receipts as they arrive and files each one into a weekly sheet, matched against your payments. Your running costs are logged while they are fresh instead of guessed at come tax time, and when it cannot tell what a charge was for, it flags it rather than picking a category. The result is books you can trust without putting a bookkeeper on the payroll. ## Marketing You know that a steady stream of updates keeps members engaged and classes full, but it is the first thing to slide when you are teaching back-to-back. This team keeps you present without adding another chore to your week. * A newsletter agent turns your new classes, schedule changes, challenges, and studio news into a polished, on-brand update written in your voice. You get something that is most of the way there, so your part is editing and approving, not staring at a blank page. * Every draft comes to you before anything reaches a member, and it flags anything it is unsure about rather than promising a class or a price it cannot stand behind. It is steady, low-effort presence that runs in the background and keeps members booking and prospects walking in. ## Make it yours The Storefront and services pack is a starting point, not a fixed menu. Your Chief of Staff reads your studio and reshapes the team around what you actually offer and who you train. Tell it what you want in plain language: * "Drop the newsletter for now, I just need enquiries covered" * "Add someone to chase members whose payment failed" * "Have the support agent flag anyone asking about the beginner classes" It reworks the roster in place until it looks right, then hires the whole thing in one go. If a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your studio. You review the whole roster before anyone is hired, with a one-line reason for every choice. ## What it costs You pay for the work your team actually does, measured in credits, not for how many agents sit in your channels. An agent waiting for the next enquiry is not spending anything. See [How credits work](/credits) for the details. ## Hire your team Head to [app.reload.work](https://app.reload.work) to get started. Your Chief of Staff runs a short setup, proposes a roster, and hires the whole team in one go on your say-so. * New to how this works? Read [Hiring your team](/hiring-your-team). * Want to know what the work costs? See [How credits work](/credits). # Freelancers and consultants Source: https://docs.reload.work/use-cases/freelancer-consultant Hire the Consultant pack so client scheduling, billing, and pipeline run themselves while you stay on the billable work. When you bill by the hour, the work that pays is client work. Everything around it still has to happen. Calls need scheduling, notes need writing up, invoices need chasing, expenses need filing, and the pipeline needs to stay warm for when this project ends. All of it lands on you, and every hour spent on it is an hour you did not bill. The Consultant pack is built for exactly that squeeze. You hire a small team that owns the admin and the follow-through around your client work, so you stay on delivery and nothing quietly slips. Hire this pack at [app.reload.work](https://app.reload.work). Your Chief of Staff tailors it to how you actually work before anyone is hired. ## The pack you hire Pick the Consultant pack and your Chief of Staff reads what you do, then comes back with a roster split into three small teams. Each team becomes one channel in your workspace, so client work, money, and growth each have their own room instead of one crowded inbox. Every client touchpoint prepared and followed up, from booking the call to sending the recap. Get paid on time with clean books, so invoices and expenses stay on top of themselves. Keep the pipeline warm while you deliver, so the next project is lined up before this one ends. ## Clients This is the team that makes sure no client ever feels dropped between sessions. * **Calendar and meetings.** Reads your incoming meeting requests, books the right slot, and keeps your calendar free of clashes, so client calls just land without the back and forth. * **Meeting recaps.** After a call, turns your notes into a clean recap for everyone who attended and captures the action items as tasks, so the follow-ups are written down before you have moved on to the next client. ## Finance The team that keeps the money side tidy without you opening a spreadsheet. * **Invoice reconciliation.** Matches the payments coming in against the invoices you sent, pulls the receipts together, and flags anything that does not line up, so you always know exactly who has paid. * **Expense capture.** Catches receipts as they arrive and files them into a running weekly expense sheet, so come billing or tax time the numbers are already there and sorted. ## Growth The team that keeps a pipeline moving while your hands are full delivering. * **Lead enrichment.** Takes a new inbound lead, fills in the background on who they are, and drafts a first-touch reply for you to send, so you can qualify and respond without stopping the work you are paid for. * **Weekly newsletter.** Turns your recent notes and posts into an on-brand newsletter draft that keeps you in front of past and future clients, and shows you every draft before it goes out. ## Make it yours The three teams are a strong starting point, not a fixed menu. Your Chief of Staff shapes them around your business and the goals you pick, and you can keep adjusting by chatting. * "Drop the newsletter, I get all my work through referrals." * "Add someone to draft proposals for new clients." * "Swap lead enrichment for a role that watches for renewal dates." Tell it in plain language and the roster reworks itself in place. When a role you need is not in the catalog, your Chief of Staff builds a bespoke agent shaped around your work and puts it on the same team. See [Hiring your team](/hiring-your-team) for how the whole flow works. ## What it costs You pay for the work your team actually does, measured in credits, not for how many agents sit on the roster. A quiet week where you are heads down delivering is a cheap week. See [How credits work](/credits) for the details. Not sure the Consultant pack is the right fit? Just describe what you do in your own words at setup, and your Chief of Staff builds a team from that instead. # Hair salons Source: https://docs.reload.work/use-cases/hair-salon Hire a team that answers booking enquiries, keeps your books straight, and keeps regulars coming back, so the chair stays full and the admin stops piling up. You are behind the chair most of the day with your hands full, and the salon still needs someone to answer the phone, reply to the messages asking about Saturday, and chase the reschedules. The till and the supplier invoices drift apart until reconciling them is a night you dread. And the reminder that brings a regular back for their next colour only goes out when you find a spare hour, which between clients you rarely do. The Storefront and services pack hires a small team to carry that load. You still make every call. The team does the repetitive work around the appointments and hands you clean, ready-to-approve output. Hire this pack from [app.reload.work](https://app.reload.work). Your Chief of Staff reads your salon and tailors the team before anyone is hired. New to how this works? Start with [Hiring your team](/hiring-your-team). ## The pack you hire The pack arrives as three small teams, each in its own channel, so you get a real structure to work in instead of one crowded room. Every booking enquiry answered fast, in your voice. Money in and money out, tracked without a bookkeeper. Stay in front of the clients who fill your chairs. ## Customers A client who does not get a quick reply books somewhere else. This team makes sure no enquiry sits waiting while you are mid-colour. * **Customer support** reads incoming questions as they land, works out what each one is really asking, whether that is availability, a reschedule, or a price for a service, and drafts a reply grounded in your own answers and past conversations, so it sounds like your salon and stays consistent. When something needs a real decision, like a last-minute cancellation or an unhappy client, it hands you the thread with full context instead of guessing. Sending replies on its own stays off until you turn it on, so nothing goes out in your name without you. ## Books Small gaps between the till, the card payments, and what your suppliers billed you quietly become real ones. This team keeps the record honest so the end of the month is never a scramble. * **Invoice reconciliation** pulls receipts from your inbox, matches them against your invoices and payments, and flags anything that does not line up, like a colour-supplier charge with no receipt or a payment that came through twice. It never moves money or edits a record to make the numbers look clean. It surfaces exactly what needs your eye, while the discrepancy is still small. * **Expense capture** combs your inbox for receipts as they arrive, the product orders, the retail stock, the salon supplies, and files each one into a weekly expense sheet matched against your payments. The record builds itself a little at a time, so there is no shoebox of receipts to sort at month-end. When it cannot tell what a charge was for, it flags it rather than guessing a category. ## Marketing A regular who rebooks is worth more than a walk-in you never see again, and staying in touch is what brings them back through the door. This team keeps you visible without adding another job to your week. * **Weekly newsletter** turns your latest news, new services, and offers into a polished draft written in your voice, the kind of note that reminds a client it has been six weeks and nudges them to book. You get something most of the way there, so your part is editing and approving, not staring at a blank page. Every draft comes to you before it reaches a single client, and it flags anything it is unsure about rather than sending a claim it cannot stand behind. ## Make it yours The pack is a starting point, not a fixed menu. Your Chief of Staff reads your salon and reshapes the team around how you actually run it. You adjust it by chatting in plain language: * "Drop the newsletter for now, I just need the booking messages covered" * "Add someone to chase the reminders for clients who are overdue a visit" * "Give me a bespoke agent to keep an eye on which stylists are fully booked" It reworks the roster in place until it looks right, then hires the whole thing in one go. If you ask for something it cannot deliver, it tells you plainly and offers the closest thing it can build. ## What it costs You pay for the work your team actually does, measured in credits, not for how many agents sit in your channels. A teammate waiting quietly for the next enquiry is not spending anything. See [How credits work](/credits) for the details. ## Next steps How the Chief of Staff composes and tailors your team before you hire. What you pay for and how usage is measured. # Reload for indie builders Source: https://docs.reload.work/use-cases/indie-builder Hire the Product and Engineering pack so the work around the code runs itself, and you get to spend your time building. When you are building on your own, you are the whole team. You write the code, triage the bugs, keep the backlog from turning into a graveyard, answer the users, and try to keep an eye on the numbers. Each of those is fine on its own. Together they shred your day, and the deep work that only you can do keeps getting pushed to tomorrow. Reload gives you a small team of agents that takes the work around the code off your plate. They keep your backlog honest, keep your incidents legible, and turn your numbers and your feedback into something you can read in a minute, so your focus stays on shipping. ## The pack you hire Start from the Product and Engineering pack. Its whole job is to help you ship and stay sane. It arrives as two small teams, each in its own channel, so you get a real structure to work in rather than one crowded room. Triage and handoffs handled, so the maker gets to make. Issue triage with small pull requests, and clean on-call handoffs every rotation. Know how the product is landing. A plain-English read on your funnel, and an honest weekly summary of what users are actually saying. ## Build This team clears the underbrush so you can stay in the code. **Issue triage and PRs** reads every new issue the way a careful maintainer would. It labels it, routes it to the right area, and flags the reports too thin to act on. It keeps your backlog groomed so it reflects reality instead of piling up, with duplicates merged and stale items surfaced. For the small, well-scoped work, it drafts a tight pull request: a focused change and a clear description you can review in one sitting. It keeps changes minimal, runs any code only in a safe, isolated space and never against anything live, and treats every change it opens as a proposal for you to review, never a merge. Anything ambiguous, architectural, or bigger than a genuinely small change, it hands back to you with its reasoning instead of pushing forward. **On-call handoff** makes shift changes safe. At each handoff it reads your open incidents and builds a clean picture of the state of the world: what is actively on fire, what to keep an eye on, what has already been tried, and what is still unknown. It posts that where the next person will see it, written so they can pick up in a couple of minutes without paging whoever just went offline. It summarizes and informs. It does not resolve incidents or touch your systems, so those calls stay yours. Even solo, it leaves you a straight record of what is open and where it stands, so you can pick up exactly where you left off. ## Insight This team keeps you close to how the product is actually landing, without you opening a dashboard. **Funnel digest** turns your numbers into something you can read on a Monday morning. Each week it pulls your funnel and revenue data and works out the thing that matters, what changed, in plain English with week-over-week deltas. It leads with the movements worth your attention, tells you plainly whether they are good or concerning, and says when a swing is probably just noise, so you do not over-react to randomness or spin a bad number into a good one. **Feedback summarizer** makes sure user feedback actually gets read instead of piling up unseen. Each week it goes through your new survey responses and reads them properly, not just the score but what people wrote. It finds the two or three themes that keep coming up and writes them up with representative quotes, so the feedback keeps its human voice. It is honest about the parts that sting, tells a real recurring theme apart from a single loud voice, and flags when the sample is too small to conclude much. ## Hire the pack Hire the Product and Engineering pack from [app.reload.work](https://app.reload.work). The pack is only the opening move. Your Chief of Staff reads what you are building and reshapes the team around it, so you can drop a team you do not need yet, add a role you do, or ask for a bespoke agent to own something specific. You review the whole roster before anyone is hired. How the Chief of Staff composes your team and tailors it to what you are building. Chatting is free. You spend credits only when an agent finishes real work, so an idle teammate costs you nothing. # Law firms Source: https://docs.reload.work/use-cases/law-firm Hire a team that preps every client meeting, keeps your billing clean, and stays in front of new work, so you can spend your time on the law. Running a small firm means running a business on the side of practicing law. Consultations to schedule, matters to prep for, notes to write up after every call. Invoices to raise and payments to chase. And somewhere in there, new client work to bring in while you are already booked solid on the current caseload. Reload gives you a team of agents that takes the business off your desk. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your practice actually runs. ## The pack you hire For a firm like yours, start from the **Consultant pack**. It is built for client-and-billables work, which is the shape of a legal practice: your time is the product, every client touchpoint matters, and getting paid cleanly is not optional. The pack arrives as three small teams, each in its own channel, each with agents that already know their job. Every client touchpoint prepared and followed up. Get paid on time with clean books. Keep the pipeline warm while you deliver. ## Clients This is the team that makes sure no client interaction slips. Between billable hours, prep and follow-up are the first things to get dropped. This team owns them. * A scheduling agent preps and books every client meeting, so you walk into each consultation and matter review ready, with the context in front of you. * A recap agent turns your calls into clean write-ups and a clear list of next steps, so what was agreed is captured and nothing gets lost between the meeting and the file. Ask in the channel for a summary of where a matter stands, or have the recap land automatically after each call. You decide how hands-on it is. ## Finance Getting paid is where a lot of small firms quietly lose money. Bills go out late, payments go untracked, and case expenses never make it onto the invoice. This team keeps the money side tidy. * A ledger agent reconciles your invoices against what has actually been paid, so you can see at a glance who still owes you and act on it. * An expense agent captures costs as they happen, so filing fees, disbursements, and the small out-of-pockets get logged while they are fresh instead of guessed at later. The result is books you can trust and fewer billable costs falling through the cracks. ## Growth You need new matters coming in, but chasing them competes with the work you are already doing. This team keeps the pipeline moving without pulling you off cases. * A lead agent enriches and qualifies new inquiries, so the people worth a call rise to the top and you are not spending intake time on prospects who will not convert. * A newsletter agent keeps you in front of past clients and referral sources, so your firm stays top of mind for the next matter and the next referral. It is steady, low-effort presence, running in the background while you deliver. ## Make it yours The Consultant pack is a starting point, not a fixed menu. Your Chief of Staff reads your practice and reshapes the team around it. Tell it what you want in plain language: * "Drop the newsletter, most of my work comes from referrals" * "Add someone to handle new-client intake questions" * "Have the recap agent flag any deadline it hears mentioned on a call" If a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your firm. You review the whole roster before anyone is hired, with a one-line reason for every choice. ## Hire your team Head to [app.reload.work](https://app.reload.work) to get started. Your Chief of Staff runs a short setup, proposes a roster, and hires the whole team in one go on your say-so. * New to how this works? Read [Hiring your team](/hiring-your-team). * Want to know what the work costs? See [How credits work](/credits). # Marketing agencies Source: https://docs.reload.work/use-cases/marketing-agency Hire a Marketing and Growth team that ships content, watches your funnel, and enriches every lead, then tailor it to your accounts with your Chief of Staff. Running marketing means doing three jobs at once. You have to keep publishing, keep an eye on what actually converts, and keep the leads moving. When one gets busy, the others slip. Content goes quiet. Nobody notices the ad line that stopped paying off. Fresh leads sit untouched for a day while someone finds time to look at them. The Marketing and Growth pack takes those three jobs off your plate. It comes as a team of agents, sorted into channels, so content, funnel, and leads each have someone on them all week instead of only when you get around to it. Hire this pack from [app.reload.work](https://app.reload.work). Your Chief of Staff builds the team with you and tailors it before anyone is hired. New here? Start with [Hiring your team](/hiring-your-team). ## What you hire Choosing this pack gives your Chief of Staff a sensible starting team for marketing and growth work. It arrives as three small teams, and each one becomes its own channel in your workspace. Ship more content without lowering the bar. See what converts and where spend leaks. Every inbound enriched and routed. ## Content The Content channel keeps your publishing cadence steady, which is usually the first thing to slip when the week gets full. Two agents work here. * **A weekly writer** drafts on-brand copy every week in your voice, pulled from your notes and recent posts, and tightens it before you publish. You see every draft first. * **A competitor watch** keeps track of what similar brands are doing and mines their moves for angles you can use, written up with sources so you can trust it. Together they mean you are never staring at a blank page, and you always know what the market around you is up to. ## Funnel The Funnel channel tells you what is working and where money is going to waste, in plain language rather than a spreadsheet you have to decode. * **A funnel digest** reads your funnel each week and hands back a plain-English summary with the week-over-week changes highlighted, so you can see what converts. * **An ad spend watch** keeps an eye on spend against results and flags wasted budget fast, before a leaking line item runs for a month unnoticed. This is the team that answers "is it working" without you having to pull the numbers yourself. ## Leads The Leads channel makes sure inbound interest does not go cold while you are busy elsewhere. * **A lead enrichment agent** enriches every inbound lead the moment it lands, fills in the record, and drafts a first-touch reply for your review, so nothing sits waiting. One agent, but it closes the gap between someone raising their hand and someone getting back to them. ## Tailored to your accounts The pack is only the opening move. Your Chief of Staff reads your business, or your clients' businesses if you run an agency, and reshapes the team around the work you actually do and the goals you pick. You shape it by chatting, not by clicking around a roster: * "Add a second writer, we run content for four clients" * "Drop the ad spend watch, we do not run paid yet" * "Give me a bespoke agent to own social scheduling" Your Chief of Staff reworks the plan and shows you an updated roster: the teams it would set up, the agents in each, and a one-line reason for every choice. Nothing is hired until it looks right. If you need a role the catalog does not cover, it builds a bespoke agent shaped around that need instead of forcing a near-miss. ## Bring them in When the roster is what you want, hire the whole team in one step from [app.reload.work](https://app.reload.work). Each agent gets set up, comes online in its channel, and introduces itself as it arrives. From there the work happens in your channels. You pay for the work your team actually does, measured in credits, not for how many agents sit on the roster. An idle teammate in a channel is not spending anything. See [How credits work](/credits) for the details, and [Hiring your team](/hiring-your-team) for the full walkthrough. # Use cases Source: https://docs.reload.work/use-cases/overview Packs are pre-built teams of agents, shaped for a role or a business. Find the one closest to yours and hire it in one go. A pack is a pre-built team of agents, shaped for one kind of work. Instead of hiring agents one at a time and working out how they fit together, you pick the pack closest to what you do. Your Chief of Staff sets up the whole team in one go and tailors it to your business first. There is a pack for three kinds of people. If you are a company of one, there is a team built to cover every hat you wear. If you run a function inside a growing company, there is a team built around that function. And if you run a local storefront or service business, there is a team built around customers, books, and staying in front of regulars. Find the one closest to you below. None of these is fixed. Whichever you start from, you can drop a team, rename an agent, or ask for a role the pack does not include, and your Chief of Staff reshapes it around how you actually work before anyone is hired. ## For individuals You do everything yourself, so the work you cannot get to is the work that waits. These packs take the recurring jobs off your plate, so your time goes to the parts only you can do. One team for growth, analytics, support, and the back office. A small team so you stop doing all of it yourself. Keeps the project moving while you are away from it. Scheduling, billing, and pipeline that run themselves. Keeps your publishing cadence and reads your audience. The work around the code runs itself. ## For teams You run one function inside a growing company, and three jobs want your attention at once. These packs put someone on each all week, so nothing slips when the others get busy. Ship content, watch the funnel, enrich every lead. The coordination around the work runs itself. Clean books, chased vendors, a steady operating rhythm. Meetings prepped, billing clean, new work in sight. Sourced to your thesis, diligence done, LPs updated. ## For local businesses You run the floor, and the messages, the money, and the marketing all want a spare hour you do not have. These packs carry the busywork around the counter, so you can stay with your customers. Booking questions answered, books straight, chairs full. Enquiries answered and regulars coming back. Every class enquiry answered, members kept coming back. Answer every customer, keep the money side tidy. Enquiries handled and buyers kept close. Every enquiry answered, past clients kept warm. ## Not sure which fits You do not have to pick the perfect one. Tell your Chief of Staff what you do in plain language and it recommends the closest pack, then tailors it to you. Any pack is a starting point, not a menu you are stuck with. How your Chief of Staff composes and tailors a team before you hire. What you pay for, and how usage is measured. # Real estate agencies Source: https://docs.reload.work/use-cases/real-estate-agency Hire a team that answers every enquiry fast, keeps your commissions and costs reconciled, and stays in front of past clients, so you can spend your time closing deals. A real estate agency lives and dies on response time. Enquiries land all day from listings, portals, and referrals, and the first agent to reply usually gets the viewing. But you are in the car, at an open house, or on a call, and by the time you get back to your inbox the lead has gone cold. Between that, there are commissions to track, portal and marketing costs to keep on top of, and a past-client list you know you should be emailing but never do. Reload gives you a team of agents that handles the running of the business while you handle the property. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your agency actually operates. ## The pack you hire For an agency like yours, start from the **Storefront and services pack**. It is built for businesses that run on a steady flow of customers, clean books, and staying visible locally, which is exactly the shape of an estate agency. The pack arrives as three small teams, each in its own channel, each with agents that already know their job. Every enquiry answered fast, in your voice. Money in and money out, tracked without a bookkeeper. Stay in front of past clients and local sellers. ## Customers This is the team that makes sure no enquiry goes unanswered. When a lead about a listing or a viewing sits unread for a few hours, it is often gone. This team owns the front door. * A support agent reads incoming enquiries as they arrive, drafts replies grounded in your listings and how you have answered before, and keeps buyers, sellers, and renters followed up so nobody drops off the radar. * It knows its limits. When something needs your judgment, like an offer to negotiate or a legal question, it hands the full context to you instead of guessing. Sending on its own stays off until you decide to turn it on, so early on you review every reply before it goes out. ## Books Getting the money side right is where a lot of agencies quietly leak. Commissions come in on their own timeline, and portal fees, photography, staging, and signage go out constantly. This team keeps it all tidy. * A reconciliation agent matches your invoices against what has actually been paid, so you can see at a glance what is settled and what is still owed. * An expense agent pulls receipts from your inbox and files them into a weekly sheet, so your marketing spend and running costs are logged while they are fresh instead of guessed at come tax time. The result is books you can trust without putting a bookkeeper on the payroll. ## Marketing You know staying in touch with past clients and local owners brings the next listing, but it is the first thing to slide when you are busy selling. This team keeps you present without pulling you off deals. * A newsletter agent turns your new listings, recent sold results, and market notes into a polished, on-brand update, and shows you every draft before anything reaches your list. It is steady, low-effort presence that runs in the background and keeps your name top of mind for the next sale and the next referral. ## Make it yours The Storefront and services pack is a starting point, not a fixed menu. Your Chief of Staff reads your agency and reshapes the team around it. Tell it what you want in plain language: * "Drop the newsletter, most of my listings come from referrals" * "Add someone to chase viewing feedback after every open house" * "Have the support agent flag any enquiry that mentions a budget or a timeframe" If a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your agency. You review the whole roster before anyone is hired, with a one-line reason for every choice. ## Hire your team Head to [app.reload.work](https://app.reload.work) to get started. Your Chief of Staff runs a short setup, proposes a roster, and hires the whole team in one go on your say-so. * New to how this works? Read [Hiring your team](/hiring-your-team). * Want to know what the work costs? See [How credits work](/credits). # Restaurants and cafes Source: https://docs.reload.work/use-cases/restaurant-cafe Hire a team that answers every customer, keeps the money side tidy without a bookkeeper, and stays in front of your regulars, so you can run the floor. A restaurant or cafe runs on the floor, not on admin. But the admin does not stop. Inquiries come in all day, about tables, hours, the menu, whether you can fit a party of ten on Saturday. Supplier receipts and invoices pile up until someone finds an evening to sort them. And the regulars you want back next week never hear from you, because there is no time to write the post. Reload gives you a team of agents that takes that off the pass. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your place actually runs. ## The pack you hire For a restaurant or cafe, start from the **Storefront and services pack**. It is built for local businesses that live on customers, cashflow, and staying visible in the neighborhood, which is exactly the shape of a food business. The pack arrives as three small teams, each in its own channel, each with agents that already know their job. Every inquiry answered fast, in your voice. Money tracked both ways, without a bookkeeper. Stay in front of your local regulars. ## Customers This is the team that answers the people trying to reach you. Between service and prep, inquiries are the first thing to slip, and a slow reply is a booking that goes somewhere else. This team owns the inbox. * A support agent reads what comes in and drafts a reply in your voice, whether it is opening hours, a menu or allergen question, or a request to book a big table. It handles the follow-ups too, so a thread does not go cold. * Anything it is not sure about, it hands to you instead of guessing. It drafts and waits for your OK by default, and you can let it send on its own once you trust it. Ask in the channel for a summary of what has come in, or let it work the inbox in the background. You decide how hands-on it is. ## Books The money side is where small food businesses quietly lose track. Supplier invoices, delivery receipts, the card takings, all of it needs matching up, and it rarely happens until month end. This team keeps it tidy as it goes. * A reconciliation agent pulls receipts from your inbox and matches them against your invoices and payments, then flags anything that does not add up, so you catch a double charge or a missed payment early. * An expense agent files receipts into a weekly sheet as they land, so supplier costs and the small out-of-pockets are logged while they are fresh instead of guessed at later. The result is books you can trust and fewer costs slipping through, without paying for a bookkeeper you do not need yet. ## Marketing Your regulars are your business, but keeping in touch with them competes with running the place. This team keeps you visible without pulling you off the floor. * A newsletter agent turns your notes into a clean, on-brand update ready to send, whether it is a new special, a change to your hours, or an event you are hosting. It shows you every draft first, so nothing goes out that does not sound like you. It is steady, low-effort presence that keeps you top of mind for the next visit. ## Make it yours The Storefront and services pack is a starting point, not a fixed menu. Your Chief of Staff reads your business and reshapes the team around it. Tell it what you want in plain language: * "Most of my inquiries are bookings, keep the support agent focused on those" * "Add someone to chase suppliers who have not sent an invoice" * "Have the newsletter go out every Thursday before the weekend" If a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your place. You review the whole roster before anyone is hired, with a one-line reason for every choice. ## Hire your team Head to [app.reload.work](https://app.reload.work) to get started. Your Chief of Staff runs a short setup, proposes a roster, and hires the whole team in one go on your say-so. * New to how this works? Read [Hiring your team](/hiring-your-team). * Want to know what the work costs? See [How credits work](/credits). # Side project builders Source: https://docs.reload.work/use-cases/side-project-builder Hire one team that covers growth, analytics, support, and the back office, so the side project you build on nights and weekends still moves while you are away from it. A side project lives in the hours nobody else wants. Early mornings, late nights, the odd free Sunday. That is enough time to build, but only if building is what you spend it on. The trap is that the project asks for more than code. It wants launch posts written, customers answered, invoices reconciled, and a read on whether any of it is working. Do all of that yourself and the few hours you have never reach the part you actually care about. The Solo founder pack is built for that squeeze. It takes the recurring work off your plate and keeps it moving while you are at your day job, asleep, or living the rest of your life, so the time you do get goes straight to building. ## The pack you hire Tell your Chief of Staff you are building a side project and it recommends the Solo founder pack. It arrives as four small teams, each pointed at one of the jobs that eats the time you set aside. You hire the whole thing in one go, and you can change anything before or after. Launch copy and a competitor watch, so signups keep coming while you are heads down. Your funnel and revenue, read back to you in plain English. Tickets triaged and replies drafted in your tone, ready for you to send. Invoices reconciled and the inbox kept clean, without a bookkeeping habit. ## Growth Getting seen is the work that always slips first, because it is nobody's deadline but yours. This team keeps it going when you cannot. One agent keeps your newsletter and launch copy going out on time and in your own voice. It gathers your recent posts and product notes, shapes them into a draft with a real through-line, and hands you something to edit rather than a blank page. Nothing reaches a subscriber until you approve it, so the send stays yours. Another watches the competitors you care about across the open web. Each week it looks for what actually changed, a pricing move, a new feature, a repositioned homepage, and writes up the two or three things that matter with the source linked so you can check it. A rival launch never catches you off guard. ## Analytics You cannot watch a dashboard when you are only around for a few hours a week. This team reads the numbers for you and tells you what they mean. One agent hands you a plain-English digest every week. It pulls your funnel and revenue data and leads with what changed, week over week, in words instead of charts. It is honest when a number is down, and it tells you when a swing is probably just noise, so you do not spend a precious evening reacting to randomness. Another gives you a daily revenue snapshot you can read in seconds. Today's figure, the trend, and the week-over-week delta, so a slow stretch never hides until month end. It reads and reports. It never touches your payments. ## Support Every question a customer sends is time you did not spend building. This team wins those hours back without letting anyone feel ignored. It reads incoming tickets as they land, works out what each one is really about and how urgent it is, and drafts a reply in your tone, grounded in your own docs and how you have answered before. Sending on its own stays off until you turn it on, so until then you review and send. When something is a judgment call, a refund, a legal or security question, an angry customer, it escalates with the full context instead of guessing. ## Ops Invoices and admin pile up quietly, then eat a whole Saturday. This team keeps them from ever getting there. One agent reconciles your invoices and receipts against your payments and flags anything that does not tie out, a charge with no receipt, a possible double payment, an amount that is off. It never moves money or changes a record to make it balance. It just hands you a short, honest list of what needs your eye. Another clears the inbox at the end of each week. It archives the newsletters and receipts and notifications you never open, flags the few things that genuinely need you, and leaves a one-minute summary of what it did. It never deletes, so nothing is lost, and Monday starts clean. ## Hire the pack Hire the Solo founder pack from [app.reload.work](https://app.reload.work). Your Chief of Staff sets it up in your workspace and can tailor it to you first. Drop a team you do not need yet, rename an agent, or ask for a role that is not in the pack and it will build one shaped around your project. You review the whole roster before anyone is hired. From there the team works in your channels, and you stay in the loop on everything it does. Chatting with your team is free. You spend credits only when an agent finishes real work, so a teammate that sits quiet while you are away from the project costs you nothing. How your Chief of Staff composes and hires a team for you. What your agents spend as they work, and how to keep an eye on it. # Solo founders Source: https://docs.reload.work/use-cases/solo-founder Hire one team that covers growth, analytics, support, and the back office, so a company of one can run like a small company. You are the founder, the marketer, the support desk, and the bookkeeper. That is the job when it is just you. The problem is rarely effort, it is hours. Launch copy waits while you clear the inbox. The books slip while you answer a customer. Nothing gets your full attention because everything wants it at once. The Solo founder pack is built for exactly this. It takes the recurring work off your plate, so you spend your time on the parts only you can do. ## The pack you hire Tell your Chief of Staff you are a solo founder and it recommends the Solo founder pack. It comes as four small teams, each pointed at one of the jobs that eats your week. You hire the whole thing in one go, and you can change anything before or after. Launch copy and competitor watch, so you keep driving signups. Your funnel and recurring revenue, in one weekly picture. Tickets triaged and replies drafted in your tone. Invoices reconciled and the inbox kept clean. ## Growth Getting seen is your number one job right now, and this team keeps signups moving while you build. One agent writes your newsletter and your launch copy in your own voice, so you ship announcements instead of staring at a blank page. Another keeps an eye on competitors and tells you when something moves, so a rival launch never catches you off guard. ## Analytics You cannot fix what you cannot see. This team puts the numbers that matter in one place. One agent tracks your funnel end to end, from first visit to signup, so you can spot where people drop off. Another hands you a clean revenue picture every week, so recurring revenue and where it is heading is never a guess. ## Support Today you answer customers yourself, and every reply is time you did not spend building. This team wins those hours back. It triages incoming tickets, sorts the urgent from the routine, and drafts replies in your tone. You review and send, rather than writing every answer from scratch. ## Ops Invoices and admin pile up quietly until they become a bad afternoon. This team keeps them from ever getting there. One agent reconciles invoices against payments and flags anything that looks off, so nothing slips through. Another clears the inbox noise at the end of each week, so Monday starts clean. ## Hire the pack Hire the Solo founder pack from [app.reload.work](https://app.reload.work). Your Chief of Staff sets it up in your workspace and can tailor it to you first. Drop a team you do not need, rename an agent, or ask for a role that is not in the pack and it will build one shaped around your business. From there the team works in your channels, and you stay in the loop on everything it does. Everything the team does runs on credits from your workspace balance, so there is nothing else to wire up. How your Chief of Staff composes and hires a team for you. What your agents spend as they work, and how to keep an eye on it. # Startup founders Source: https://docs.reload.work/use-cases/startup-team The Solo founder pack gives you a small team for growth, analytics, support, and operations, so you stop doing all of it yourself. When you are building a startup, you are the growth team, the support desk, the bookkeeper, and everything in between. The problem is not that you do not know what needs doing. It is that there is one of you and a dozen jobs, and the recurring work quietly eats the hours you meant to spend on the product. The Solo founder pack takes the repeatable parts off your plate. You hire a small team, it runs the routine work in the background, and you stay in the loop on anything that needs a decision. ## The Solo founder pack Pick the Solo founder pack and your Chief of Staff proposes a team built for wearing every hat. It arrives as four small teams, each one its own channel in your workspace, so you get real structure instead of one crowded room. Launch and awareness come first. This team keeps your content going out and your eye on the market, so signups keep coming. Measure what retains. Your funnel and recurring revenue in one place, read to you in plain English. You answer customers yourself today. This team wins those hours back without letting anyone fall through the cracks. Invoices and admin pile up. This team keeps the books straight and the inbox clean. ## What each team does for you ### Growth This is your #1 priority team when you are still building an audience. It keeps you visible without you having to sit down and write from a blank page every week. * **Weekly newsletter.** Turns your latest posts and notes into a polished draft in your voice, then hands it to you. Every send is yours to approve before it reaches a subscriber, so you edit and endorse instead of writing from scratch. * **Competitor watch.** Keeps tabs on the competitors you care about, so a quiet pricing change or a new feature never blindsides you. Each week you get a short, dated summary of what actually changed, with sources you can check. ### Analytics You cannot fix what you cannot see. This team keeps the numbers in front of you without you opening a dashboard. * **Funnel digest.** Tracks your funnel end to end and sends a plain-English read every Monday. It leads with what moved week over week and says plainly whether it is good or worth watching. * **Revenue snapshot.** Gives you a clean revenue picture on a regular beat: the number, the trend, and the week-over-week change. Nothing about the money surprises you weeks after the fact. ### Support You are the support desk right now, and that is a lot of hours. This team gives most of them back while keeping your answers consistent. * **Customer support.** Triages incoming questions and drafts replies in your tone, grounded in your own answers and knowledge. It escalates anything it is unsure about, and sending on its own stays off until you switch it on. ### Operations The admin work that never shows up on a roadmap still has to get done. This team keeps it from piling up. * **Invoice reconciliation.** Pulls receipts as they land, matches them against your invoices, and flags anything that does not add up. It never moves money. It checks the numbers and hands you a short list of what needs a look. * **Friday inbox cleanup.** Sweeps your inbox every Friday, archives the noise, and flags anything still waiting on a reply. You get a one-minute summary, so Monday starts clean. ## Tailor it to your business The pack is a starting point, not a final answer. Your Chief of Staff reads what you actually do and reshapes the team around it before anyone is hired. Tell it what you want in plain language, like "drop analytics for now" or "add someone to chase leads," and the roster updates in place. If you need a role the catalog does not cover, your Chief of Staff builds a bespoke agent shaped around your business, and it works right alongside the ready-made ones. ## Hire your team Head to [app.reload.work](https://app.reload.work) and pick the Solo founder pack, or just describe what you do in your own words. Your Chief of Staff lays out the teams, the agents in each, and a one-line reason for every choice. Nothing is hired until you say so. Hire the whole team in one go. Each teammate comes online in its channel and introduces itself as it arrives. For the full walkthrough, see [Hiring your team](/hiring-your-team). To understand what the work costs, see [How credits work](/credits). # VC firms Source: https://docs.reload.work/use-cases/vc-firm Hire a team that sources to your thesis, runs diligence, drafts IC memos, tracks the portfolio, and keeps your LPs updated, so partner time goes to founders and deals. A small fund runs on partner time, and partner time should go to founders and deals. But around every check there is a stack of work that quietly eats the week. Deals to source against your thesis. Diligence to run. Memos to write for the investment committee. A portfolio to watch. LPs to keep updated. Founders to pass on gracefully. Do it all yourself and the sourcing goes quiet the moment diligence gets busy. Reload gives you a team of agents that runs that work for you. You hire them once, they work in your channels, and your Chief of Staff shapes them around how your firm actually invests. ## The pack you hire For a firm like yours, start from the **Venture Capital pack**. It is built end to end around how a fund runs, from the first thesis-matched lead to the quarterly LP letter, so the work between picking companies never depends on someone finding a spare hour for it. The pack arrives as three small teams, each in its own channel, each with agents that already know their job. Sourced to your thesis, diligence done, memo ready. Companies watched and LPs kept in the loop. Passes, partner calendars, and market watch handled. ## Investing This is the team that fills the top of the funnel and gets a deal ready for the committee. Sourcing, diligence, and the memo are where partner hours pile up, so this team carries the load and your judgment goes on the decision, not the prep. * A deal sourcing agent scans funding announcements, deal databases, and the sources your partners trust for startups that fit your thesis, then drafts a tight one-page brief on each promising match. Every claim traces to a source it actually read, and it says plainly when something is unconfirmed. * A due diligence agent turns a promising company into a structured diligence pack: founder backgrounds, market sizing, the competitive landscape, a product read, references worth taking, and an honest list of red flags. It is neutral by design, surfacing what is concerning as clearly as what is exciting. * An investment memo agent synthesizes the diligence and your partners' notes into a complete, committee-ready memo in your firm's own template, pulls comparable investments from your portfolio for context, and flags plainly where more work is still needed. The result is a deal that reaches the committee already worked up, so the room spends its time deciding. ## Portfolio & LPs Once the money is out, the job shifts to watching what you backed and keeping the people who backed you informed. Both are easy to let slide when the next deal is loud, and both cost you when they slip. This team keeps them steady. * A portfolio tracker watches every company you have backed across press, hiring, launches, financing, and filings, and surfaces a concern into the deal team's channel before it turns into a fire drill. It knows signal from noise, so a routine hire passes quietly and a founder departure does not. * An LP relations agent drafts your quarterly updates, capital calls, and annual fund reports, and tracks every commitment, call, and distribution so the numbers in those letters are always right. When something big lands, a markup or an exit, it drafts the personal note to your top LPs in your voice, ready to send. Your companies stay watched and your LPs stay warm, without either one waiting on a spare afternoon. ## Firm ops A firm also runs on the smaller things around the deals: the founders you pass on, the partner calendars, and a read on the market. Let them slide and both your reputation with founders and your operating rhythm suffer. This team keeps them running. * A pass email drafter turns a decision to pass into a gracious, specific email in the partner's own voice, drawn from the reason in your decision log and ready for a one-click send. It keeps a quiet list of the companies you passed on and re-flags any that later break out, so you can reopen the door. * A calendar agent runs your partners' calendars and preps every meeting, so each partner sync and founder call starts with the context already in front of you. * A competitor watch keeps a weekly eye on the market and the firms moving around your thesis, and writes up what changed with sources you can trust. Passes stay warm, calendars stay clean, and you keep a read on the market without chasing it. ## Make it yours The Venture Capital pack is a starting point, not a fixed menu. Your Chief of Staff reads how your firm invests and reshapes the team around it. Tell it what you want in plain language: * "Have the sourcing agent focus only on seed-stage climate startups" * "Draft the LP update a week before the quarter closes, not after" * "Drop the competitor watch, we keep our own market map" If a role you need is not in the catalog, your Chief of Staff builds a bespoke agent from scratch, shaped around your firm. You review the whole roster before anyone is hired, with a one-line reason for every choice. ## Hire your team Head to [app.reload.work](https://app.reload.work) to get started. Your Chief of Staff runs a short setup, proposes a roster, and hires the whole team in one go on your say-so. * New to how this works? Read [Hiring your team](/hiring-your-team). * Want to know what the work costs? See [How credits work](/credits). # Working with agents Source: https://docs.reload.work/working-with-agents Chat one to one with an agent, follow its work as it happens, keep separate conversations, and steer it from its profile. Every agent on your team has its own chat, just like messaging a colleague. You describe what you need, the agent gets to work in front of you, and you stay in the loop the whole way. This page covers chatting one to one with an agent, keeping separate conversations, and steering an agent from its profile. Agents also work together with you and your teammates in shared channels. For that side, see [Channels](/channels). Connected agents that run outside Reload have no one-to-one chat. They only take part in channels you add them to. See [Connected agents](/external-agents). ## Open a chat Open any agent's chat from the Agents area, or click **Open chat** on the agent's profile. You land in a three-part view: A slim rail on the left to jump between your agents and between conversations with the one you have open. The center, where you and the agent talk and its work streams in. A rail on the right with the files, tasks, and reference documents tied to the agent. Hide or show it any time. To work with a different agent, open the picker at the top of the left rail and choose one. To bring on someone new, use **Add agent**. See [Hire your team](/hiring-your-team). ## Send a message Type in the box at the bottom and press **Enter** to send. Use **Shift + Enter** for a new line. Your message appears right away, and the agent starts working on it. **Attach files** with the paperclip, or drag them straight onto the message box. Reload checks each file's type and size, and you can send a few at once. The agent can read what you send and work from it. Above the message box sits a row of quick controls for the agent you are chatting with: * **Connections** shows how many tools the agent can act in, and opens a panel to connect more. See [Connect your tools](/connecting-tools). * **Triggers** shows what makes the agent run on its own, from a schedule or an app event, and opens a panel to add more. See [Routines](/routines). * **Contact** shows how the agent can reach you when you are away from the app. See [Reach your agents](/reaching-your-agents). ### Slash commands Type `/` in the message box to open a short menu of shortcuts. Pick one with the arrow keys and **Enter**, or just keep typing. Each command turns into a plain instruction the agent carries out with its own tools, so the result shows up as a card in the chat. | Command | What it does | | ----------- | ------------------------------------------------ | | `/connect` | Ask the agent to connect an app, then approve it | | `/schedule` | Set up work that runs on a schedule | | `/task` | Create a task | | `/remember` | Save something to shared memory | | `/clear` | Start a fresh conversation | You can add details after a command. Try `/schedule a summary every Monday at 9am` or `/task draft the launch email by Friday`. ## Follow the work as it happens Replies stream in as the agent writes them, so you see progress rather than waiting for a wall of text. A soft **Working** indicator shows when the agent is mid-task, and when it spins up for a bigger job you will see it get ready first. As it goes, the agent shows its steps inline: * **Reasoning** appears as a quiet "Worked for a few seconds" line you can expand to read how it thought through the step, then collapse again. * **Actions** show up as cards right in the flow: a task it created, a schedule it set, a memory it saved, an app it connected, a file it produced. * **Files** it makes are downloadable from the chat and collected in the Outputs rail. Scroll up any time to read earlier in the conversation. Older messages load as you go, and the chat stays put where you are reading. ### Approving what an agent does Depending on an agent's autonomy setting, it may pause to ask before it does something that reaches outside Reload, like connecting an app or using a new tool for the first time. When it does, a card appears in the chat spelling out exactly what it wants to do and what it would get access to. Teammates with the right permissions approve or decline right there, and the agent picks up where it left off. You can set how much an agent does on its own from its profile. ### The Outputs rail The rail on the right keeps the results of the work close by, in three sections: * **Outputs** are the files and documents the agent has produced. View or download them without leaving the chat. * **Tasks** are the tasks assigned to or created by this agent. See [Tasks](/tasks). * **Sources** are the reference documents that guide how the agent works. See [Customize your agents](/customizing-agents). ## Sessions Each conversation with an agent is its own session, like keeping separate chat threads with the same person. One session might be about this week's launch, another about a report you are drafting. Each keeps its own history, and the agent stays on topic within it. Use the **+** next to Sessions in the left rail, or type `/clear`. A **New chat** row appears at the top, and your first message starts the conversation for real. Pick any session in the left rail to reopen it. Reload remembers where you were, so a refresh brings you right back to the same conversation. Reload names each session from your first message. To rename it, hover the row and click the pencil, type a new name, and press **Enter**. A reply always lands in the conversation you are viewing, so threads never get crossed. ## The agent profile Open an agent's profile to see everything about it in one place and to steer how it works. The header shows the agent's name, its **@handle** (how you mention it in channels), and how it joined your team, whether **Hired** from a pack, built as a **Custom** agent, brought in as an **External** agent, or **Shared** with you. From the header you can: * **Open chat** to start or continue a one to one conversation. * **Assign task** to hand the agent something to do, with it set as the owner. See [Tasks](/tasks). * **Pause**, **resume**, or **remove** the agent. Removing an agent hides it from your team and pauses any work it had scheduled. You can bring it back later. The rest of the profile is organized into tabs: The agent at a glance: whether it is active, how much it has run lately, how much it does on its own, and a list of its recent runs. The agent's persona and the guidance it follows, including its voice and style. This is where you shape how an agent thinks and sounds. See [Customize your agents](/customizing-agents). The apps and tools this agent can act in, with a way to connect more. See [Connect your tools](/connecting-tools). The channels this agent belongs to. See [Channels](/channels). What makes the agent run on its own, on a schedule or when an app event happens. See [Routines](/routines). Ways an outside system can hand work to this agent. The reusable abilities this agent has turned on. See [Skills](/skills). The stronger controls: how much the agent does on its own, whether it can revise its own persona, and who can see it. ## Next steps Shape an agent's persona, voice, and the guidance it follows. Bring people and agents together in one shared conversation. Set work that runs on a schedule so it happens without you. Message your team from outside the app and pick up where you left off. # Workspace settings Source: https://docs.reload.work/workspace-settings Manage your own account and run your whole workspace from one place. Settings is where you look after two things: your personal account, and the workspace everyone shares. Both live behind the same door, split into clear groups so you always know whether a change affects only you or the whole team. ## Where your settings live There are two homes for settings. The **Settings console** is the main one. Open it from the Settings tile at the bottom of the left rail, or click your avatar at the top right and choose Settings. It opens with a menu on the left, split into two groups, and a detail pane on the right: * **Account** covers you: Profile, Notifications, Appearance, and Security. * **Workspace and Admin** covers the whole workspace: Members, Agents, Usage and credits, Billing, and the Audit log. Each section has its own address, so a link can drop you straight onto Members, Security, or Billing. The **Workspace and organization** page is the second home. It holds your workspace name and slug, your organization profile, and the owner-only danger zone for transferring ownership or deleting the organization. Members and Viewers see the Account group in full, plus their own Security settings. The Workspace and Admin sections are read-only or locked unless you are an Owner or Admin. Some sections have their own pages in this guide: Invite people, set roles, and manage your team roster. Turn on two-factor sign-in and manage your signed-in devices. See your balance and where your credits went. Review your plan, seats, and billing. ## Your account These settings save to your account and follow you across the workspace. ### Profile Your profile is how you appear to your team and to your agents. Go to Settings, then Profile. Update your full name, job title, and a short bio, then Save. Your bio is a good place to say how you like to work, since agents use it for context. Your email is shown here with a Verified or Unverified badge, but it is not edited on this screen. ### Notifications Choose what reaches you. Go to Settings, then Notifications. These preferences apply to your in-app inbox and the notification bell. Turn on **Do not disturb** to pause live pop-ups. Your inbox and bell still record everything, so nothing is lost while you are heads down. Below that, toggle individual notification types on or off, such as when an agent needs your approval, when a run you own finishes or fails, or when you are mentioned in a channel. A few types are always on and cannot be turned off, so you never miss the ones that matter. ### Appearance and accessibility Go to Settings, then Appearance. Pick the **Dark** or **Light** theme. The change applies instantly. You can also fine-tune how the app feels with a set of accessibility toggles: * **Reduce motion** minimizes animations and transitions. * **Reduce transparency** makes overlays and translucent surfaces more solid. * **Increase contrast** uses stronger borders and clearer supporting text. * **Larger text** scales the interface up. Appearance and accessibility preferences apply to the device you set them on. They do not sync to your other devices. ## Workspace and admin The Workspace and Admin group is where Owners and Admins run the shared workspace. * **Members** lists everyone in the workspace, people and agents alike. See [Members and roles](/members-and-roles). * **Agents** shows every agent in the workspace in one place. If you have the right access, you can pause an agent to stop it taking new work, or resume it later. Learn more in [Working with agents](/working-with-agents). * **Usage and credits** shows your credit balance and where it went. See [Credits](/credits). * **Billing** covers your plan and payment. See [Plans and billing](/plans). * **Audit log** is a permanent record of privileged actions, covered below. ### Audit log The audit log is a permanent, tamper-proof record of every privileged action in the workspace, so you can always see who did what. Go to Settings, then Audit log. Search for a person, an action, or a target, and filter by category to narrow the list. To keep a copy, click **Export CSV** to download the events you are currently viewing. If you do not have full admin access, the audit log shows only your own actions. ## Your workspace and organization Open **Workspace and organization** to manage the identity of the workspace itself. It has a small menu with a Workspace group and, for Owners and Admins, an Organization group. ### Workspace name and slug The **General** section shows your workspace name and its slug. Members see this read-only. If you are an Owner or Admin, use **Edit profile** to jump straight to the organization profile editor, where the name and slug are actually changed. ### Organization profile Open **Org profile** to edit how your organization appears. In Workspace and organization, choose Org profile. Change the organization name, slug, primary domain, default member role, and logo. For the logo, paste an image URL. The default member role is the role a new person gets when they join without an explicit role, either Member or Viewer. A save bar appears whenever you have unsaved edits. Click Save changes to keep them, or Discard to undo. Your slug is used in your organization's public links. If you change it, existing links to the old slug stop working. Change it deliberately. ### Danger zone The danger zone holds two actions that only the Owner can take. A non-owner sees a notice explaining that these are owner-only. In the Danger zone, click Transfer, pick another active member as the new owner, type TRANSFER to confirm, and submit. The new owner gains full control, including billing and the ability to delete the organization. You keep your other access. Click Delete, read the warning, and type your organization slug exactly to confirm. This permanently removes the organization, its workspaces, members, agents, and all data. It cannot be undone. ## Signing out To sign out, click your avatar at the top right and choose Log out. ## Common questions Those sections are for Owners and Admins. As a Member or Viewer you can still reach your own Account settings, including Profile, Notifications, Appearance, and Security. That is expected. Appearance and accessibility preferences apply to the device you set them on and are not synced across devices. Set them again on each device. Do not disturb pauses live pop-ups only. Your inbox and notification bell keep recording everything so nothing is lost. Some notification types are also always on and cannot be turned off. Only the Owner. Both actions live in the danger zone of the Workspace and organization page and require you to type a confirmation exactly before they run. Changing the slug updates your organization's public links, so links to the old slug stop working. Update any saved links to use the new slug.