Skip to main content
A connector gives your agents tools beyond chat: a browser, a database, Linear, Notion, Gmail. Each connector is an MCP server (Model Context Protocol, the common way to hand tools to an AI agent). Use this page to connect one, give it to the agents that need it, and take it back. Connecting and giving access are two separate steps. Connect starts the connector, but no agent can see or call its tools until you give it to that agent, on the connector’s detail pane or on the graph.

Before you begin

  • Node.js on the machine that runs Clawboo. Most connectors start with npx, and the first start downloads the connector’s pinned package.
  • For a connector you sign in to (Linear, Sentry, Stripe, Composio), a browser on that same machine. The sign-in page sends you back to a temporary address on that machine.
  • An account or key for the service, when the connector asks for one.

Connect a connector

1

Open Connectors

Click Connectors in the sidebar, or press Cmd/Ctrl + 3.
2

Find the connector

Search by name, or pick a category. Connected ones come first, then the ones that need the least setup.
3

Press Connect

What happens next depends on what the connector needs:
4

Check that it connected

The row shows a green tick, and a message confirms it with the connector’s tool count and the reminder “No agent can use it until you give it access.”
Now give it to an agent: see Give a connector to an agent. A connector that cannot run on this install shows Why not instead of Connect. It says what is missing and offers a config block you can paste into another app that can run it. Keys and sign-in tokens go to Clawboo’s encrypted vault on this machine, and no API sends them back to the browser. A key you exported in your shell (for example NOTION_TOKEN) is not picked up: enter it in Connectors.
Filesystem suggests the folder you started Clawboo from, plus your Documents, Desktop and Downloads folders, each only if it exists. SQLite suggests up to five .db, .sqlite or .sqlite3 files within two folder levels of where you started Clawboo. You can always type a path instead.
GitHub offers no sign-in that Clawboo can use, so its connector takes a personal access token. Create a fine-grained token, choose only the repositories your agents should work in, and give it Read and write on Contents, Issues and Pull requests. Paste it into the GitHub row.

What the list holds

The list reads top to bottom in four bands: The reviewed set holds 20 connectors. The community band holds a snapshot of 400 registry servers, shipped with Clawboo, so the list works offline. It stays hidden until you press Show 400 more at the foot of the list or search for something; then it shows 60 at a time (Show 60 more, then Show the last N).

Apps reached through Composio

Clawboo cannot sign in to Google, Slack or Atlassian directly, so those apps go through Composio, a separate service that signs you in to each app.
1

Add your Composio key

In the Reached through Composio band, press Get a key to open Composio’s API keys page. Copy the project key (it starts with ak_), paste it into the Composio key field and press Save. You do this once.
2

Connect an app

Press Connect on the app and approve it in the tab that opens. Back in Clawboo, the app’s row reads Connected.
3

Connect the Composio connector

Your agents reach every Composio app through the Composio connector in the Popular band. Press Connect on it once and sign in to Composio.
4

Give the app to an agent

Give the app to an agent on the graph, the same way as any connector (next section): the graph’s picker lists connected apps with the connectors. Each app an agent holds shows as its own tile, so you can see which agents can reach your email.
Composio keeps each app’s sign-in tokens on its own servers, and Clawboo holds only its own key and sign-in for Composio. Anything you connect there is reachable by Composio, not only by your agents.

Give a connector to an agent

You give access on the connector’s detail pane in Connectors, or on a graph: the Org chart, a team’s Team graph card, or the Graph card on an agent’s page. On a graph, click a Boo to open its ring of tiles; each connector tile on the ring is one the agent can use.
1

Open the connector

In Connectors, click a connected connector’s row. Its detail pane shows Who can use it? right under the connect action. It lists every agent; an agent that already has access is ticked and reads Has access.
2

Pick the agents

Tick each agent that should use it, then press Give access. Each one gets write access, and risky calls ask you first: the same access a thread drawn on the graph gives.
OpenClaw agents are listed but cannot be picked, and read OpenClaw agents cannot use shared connectors; see OpenClaw agents and connectors.
Pick Write for most connectors. Many tools do not say whether they only read, so a Read grant hides them.

Who can use a connector

  • Right after you connect it: no agent. Its tools are hidden from every agent, and a call to them is refused.
  • After you give it to an agent: that agent only, with the access you chose. Every other agent still has no access until you give it to them too.
  • The exception: the two browser connectors, which every agent gets automatically (next section).

Every agent gets a browser

The two browser connectors, Playwright and Chrome DevTools, are connected for you the first time Clawboo starts, unless that is turned off in Settings > System > Deployment (see Deployment settings), and every agent gets both with write access. When the exact version Clawboo pins is already installed, in npx’s cache or in the node_modules folder Clawboo itself is installed in, each starts from that copy rather than through npx, which keeps one fewer process running and makes no registry request. Chrome DevTools runs with its usage statistics and its check for a newer version turned off. OpenClaw agents get the tiles too, but cannot use them; see OpenClaw agents and connectors. In Chrome DevTools, actions that change something, such as opening a page or clicking, ask you first, and the two that run script in the page or hand a file from this computer to a website (evaluate_script and upload_file) count as destructive and need Admin access. Playwright works the same way with write access: an agent can open pages, click, type, fill in forms, pick options, hover, drag, press keys, manage tabs, resize the window, answer page dialogs and go back, and each of those actions asks you first. Its reading tools, such as a snapshot or a screenshot of the page, need only read access and run without asking. Four Playwright tools count as destructive and need Admin access: the two that run code (browser_run_code_unsafe and browser_evaluate) and the two that can hand a file from this computer to a website (browser_file_upload and browser_drop). To stop one action asking every time, tick Do not ask again for this for 30 days on its card (see When a connector call asks you first). Both browsers start with the Clawboo dashboard’s port blocked on localhost and every name under .localhost, on every 127.x.x.x address, ::1 and 0.0.0.0, and on the address set in HOST. Chrome applies the block when it looks up an address, so a redirect, a page’s script or a WebSocket that heads for the dashboard is stopped too. Clawboo also refuses any connector call that names one of those addresses, however it is written. Together these keep an agent from opening the dashboard to approve its own requests. They do not know a name that reaches Clawboo through a reverse proxy or one listed in CLAWBOO_ALLOWED_HOSTS; there only an access token (STUDIO_ACCESS_TOKEN) keeps an agent out, because each agent’s browser has its own profile, without your sign-in cookie. Other ports, such as an app you are developing, stay reachable. Neither browser opens files on this computer: Clawboo refuses a file: or view-source: address, or a Windows path such as C:\Users, given to either one. See What agents can do on your machine for what this does not cover. Each agent has its own browser profile, so two agents can be signed in to different accounts on the same site. To let an agent sign in without ever seeing the password, save the login under Website logins. The browser tiles sit on the agent’s ring like any other connector, so you can take them away (next section), and a removed browser stays removed after a restart.

A desktop driver for each agent’s screen

Cua Driver (category Desktop) uses the apps on an agent’s own screen: it lists windows, clicks, types and reads each app’s accessibility tree. It works on Linux only, where agents get a screen of their own (see Watch a screen and take over), and it is not given to agents automatically: connect it, then give it to the agents that need it.
  • It runs only when agents run as their own OS user (CLAWBOO_AGENT_USER), like a control link. It types into the agent’s browser, which otherwise runs as you, so an address it types there could open your files and Clawboo’s own. Without it, Connect says so, and agents keep everything else: their own screen, the live view, taking over from the dashboard, and their browser tools. Only Cua Driver waits. To turn agent isolation on, see Set it up on Linux.
  • Install it first on an agent’s Screen view (Install the desktop tools, or Install Cua Driver where agents already have a screen); Connect says so until it is installed, and refuses on macOS and Windows and where agents get no screen of their own (displays turned off, or no Xvnc). Clawboo runs only the copy it installed, never one on PATH: its rules below are checked against that release.
  • Each agent gets its own Cua Driver process, run as the agent user, on that agent’s own screen and nowhere else, with a home folder of its own inside the agent’s browser folder. It talks only to that screen’s own session bus (or to none), never to your desktop session’s, so it does not switch on accessibility, or a screen reader, on your desktop. A call that names no agent, or for an agent with no screen of its own, is refused.
  • It runs in its own process for each agent, never through a shared Cua Driver service, with its usage telemetry and its check for a newer version turned off. To turn them on, set CUA_DRIVER_RS_TELEMETRY_ENABLED or CUA_DRIVER_RS_UPDATE_CHECK in the server’s environment.
  • While a person has taken over an agent’s screen, its Cua Driver calls are refused, like its browser calls.
  • Its actions ask you first, like a browser’s. The tools that run code, start or stop programs, install things, change the driver’s own settings, or move a file between the computer and a website (run_script, run_steps, replay_trajectory, launch_app, kill_app, install_extension, install_ffmpeg, check_for_update, set_config, browser_set_input_files, browser_download, clipboard_write, and browser_prepare, which starts a browser of the driver’s own without Clawboo’s checks) count as destructive and need Admin access.
  • Clawboo refuses screenshot_out_file and output_dir, the arguments with which it writes a file wherever they point (so recordings are not available); a screenshot comes back in the reply instead.

See and take back access

The graph is the list of who can use what: every tile on a Boo’s ring is something that agent can reach. A connector’s detail pane also marks the agents that have it as Has access, under Who can use it?. A tile that several agents share shows a small count, such as x3; hover it to highlight the others. Click a connector tile to see its toolbar: In Connectors, Turn off appears when you hover a connected row. A connector you signed in to also has Sign out in its detail pane (click the row): it deletes the stored tokens and stops the connection. A server you added yourself has Remove.

When a connector call asks you first

Every call an agent makes through a connector passes Clawboo’s checks first. When a call needs your answer, an approval card appears in Needs you on the board, in the chat where the agent is working, and in the Governance and Capabilities panels. If nobody answers within five minutes, the call is refused. See Approvals for the cards themselves. Which calls ask depends on the connector:
  • GitHub, Linear, Sentry, Airtable and Notion, and any server you added yourself, ask before every call. They can read your private data, take in content from outside and send data out. That combination is how data gets leaked, so you cannot turn these questions off.
  • Actions in apps reached through Composio ask every time too.
  • On the other connectors, a call that only reads runs without asking when Clawboo reviewed the connector and the tool says it only reads. A call that changes something asks first when the connector can reach the internet. A destructive tool needs Admin access and asks every time. For Playwright, Clawboo’s own list decides which tools are destructive; on the other reviewed connectors, the connector marks them itself.
Some cards offer Do not ask again for this for 30 days. Tick it, then allow the call: for 30 days that agent calls that tool, with any arguments, without asking, under the access you gave it. A destructive call never offers the box and asks every time. Approvals remembered by an earlier version of Clawboo, when the box did not stop these prompts, are cleared once on the first start after you upgrade, so none starts allowing calls it was never meant to allow.

When an agent asks for a connector

Agents are told which connectors are connected and which ones you could add. An agent that needs one it does not have says so and stops, and a card in the chat offers a button for each connector it named, such as Connect Linear or Add key Notion. The button opens that connector in Connectors, with the asking agent already ticked under Who can use it? and marked Asked for it. As soon as the connector connects on that pane, whether by its connect button, a saved key or a sign-in, the asking agent gets write access, and risky calls ask you first. Nothing changes when the agent already has access, or when it is an OpenClaw agent, which is never ticked. If you took the agent’s access away earlier, its request does not give it back: the agent stays ticked, marked Access taken back, and Give access shares it again only if you click it. The card counts a connector as done only when it is connected and the asking agent has access. A connector that is connected but not shared with that agent shows a button such as Share Linear with Ada instead, which opens the connector and gives that agent access. Once the agent can use every connector it asked for, the card shows Continue (in a team chat): press it so the agent picks the work back up.

Add your own MCP server

At the foot of the list, Add your own MCP server adds a server that speaks MCP over stdio, which means a program Clawboo starts on this machine.
1

Fill in the form

Enter a Name, the Command (an absolute path, or a program name found on your PATH) and its Arguments, separated by spaces. Arguments go straight to the program, never through a shell.
2

Add it

Press Add. The server joins More connectors as one of your own entries.
3

Connect it and give it to an agent

Press Connect on its row, then give it to an agent under Who can use it? or on the graph.
The form has no URL field, so a remote MCP server cannot be added here. A community entry works the same way once added: Connect on it shows the exact command, package and version, and I trust it, add it makes it one of your own entries. Clawboo has reviewed neither kind, so it treats them as able to read private data, take in outside content and send data out, and every call they make asks you first.

OpenClaw agents and connectors

OpenClaw agents reach Clawboo’s tools through one setting in the OpenClaw Gateway that all of them share, so Clawboo cannot tell which OpenClaw agent is calling. As a result, connectors you give an OpenClaw agent on the graph do not reach it, the browser connectors included, and Who can use it? does not let you pick one. It can use apps reached through Composio, and each action in those apps asks you first. The Gateway’s own MCP servers and plugins are separate: they are listed in the Capabilities panel.

Keep connectors safe

A local connector is a program running under your user account. Clawboo starts it with a short list of ordinary environment variables, including npm’s prefix and cache settings and Python’s bytecode settings, plus the keys you entered for it, never your AI provider keys or npm credentials, and passes its arguments without a shell (see what connectors inherit). That is not a sandbox: the connector can still read any file your account can, including your SSH keys and Clawboo’s database. Connect only servers you would be willing to run yourself.

Troubleshooting

Connecting makes a connector available; it gives no agent access. Agents are told which connectors are connected whether or not they hold one. Give the connector to that agent under Who can use it? on its detail pane, or on the graph, then ask again.
The machine running Clawboo has no npx on its PATH. Install Node.js from nodejs.org, restart Clawboo, and press Connect again.
The first start of a connector downloads its package, which can take a minute on a slow network. The button reads Working… meanwhile. Press Connect again; the second start is faster.
The service refused the key or token. Copy it again in full, and check that it has not expired. Then open the connector’s detail pane (click its row) and paste it under What you gave it.
The key works but lacks a permission. Check the scopes or permissions on the token you created at the service, for example Read and write on Contents, Issues and Pull requests for GitHub.
Allow pop-ups for the Clawboo address in your browser, then press Connect (or Sign in) again.
You pasted the key the Composio command line uses. The key this field needs starts with ak_ and is on your Composio project’s API keys page (Get a key opens it).
Clawboo refuses that connector’s calls because its command or its list of tools changed after you gave access. If you changed its key, folder or command yourself, turn it off and connect it again in Connectors: that approves the new setup. A changed tool list is not approved again automatically, and there is no dashboard control that approves one, so read what changed in the connector’s detail pane before you rely on it.
More fixes for error messages are in Troubleshooting.

Next steps

  • Approvals: answer the cards that connector calls raise.
  • Capabilities: every skill, tool and connector your agents have, in one list.
  • Org chart: the graph where you give and take back access.
  • What agents can do: what an agent can reach on each runtime, beyond connectors.
  • Connectors API: connect connectors and manage access from scripts.
Last modified on November 16, 2026