MyClawn Computer
Server Details
Your own cloud computer run by an AI agent: signed-in browser, its own email, files, long jobs.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- 20vision/myclawn-computer-mcp
- GitHub Stars
- 0
- Server Listing
- MyClawn Computer MCP
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 9 tools
Each tool targets a clearly different action or resource, and the descriptions explain when to use each one. Minor overlap exists between activity and tasks, since both report on task progress, but the activity description explicitly covers the 'how is it going' case for the latest task while tasks lists historical task states.
All tool names use consistent lowercase snake_case with a clear myclawn_ prefix, split into desktop_ and notes_ modules. The pattern is predictable and easy to scan, even though some names are nouns and others are verbs.
Nine tools is well-scoped for a remote computer agent interface, covering running, monitoring, interrupting, files, mail, status, and memory without obvious bloat. Each tool appears to earn its place.
The surface covers the core lifecycle: start tasks, monitor them, interrupt them, send files, check sent mail, inspect status, and save/recall notes. Minor gaps include no dedicated get-task-by-id for historical task details and no update/delete operation for saved notes.
Publisher details
- Operator
- 20Vision GmbH · Publisher source
- Operator website
- https://www.myclawn.com · Publisher source
- Vendor relationship
- First-party · Publisher source
- Documentation
- https://www.myclawn.com/docs/connectors · Publisher source
- Privacy policy
- Not provided
- Trust center
- Not available
- Restrictions
- Needs a MyClawn account. A new account's first task starts a free 24-hour computer with no card; after that, tasks need a paid plan. · Publisher source
Available Tools
9 toolsmyclawn_desktop_activitySee what the computer is doingARead-onlyIdempotentInspect
Shows the latest task's steps on the MyClawn computer (pages opened, clicks, notes along the way) and whether it is still running. Use it to check on a long task, after myclawn_desktop_run reported it is still working, or when the user asks how it is going. Read-only.
Inputs: no named parameters. Returns: steps, title, running
myclawn_desktop_interruptStop the current taskADestructiveIdempotentInspect
Stops the task the MyClawn computer is running right now. Use it when the user asks to stop, or a task is clearly going wrong. The task stops at its current step; anything it already did (an email sent, a form submitted) is not undone. Safe to call when nothing is running (interrupted is then false).
Inputs: no named parameters. Returns: interrupted
myclawn_desktop_runRun a task on my MyClawn computerADestructiveInspect
Hand a job to the user's own MyClawn computer, a cloud machine with an agent on it. Use it for what you cannot do from this chat: work inside websites and portals that need the user's login (the browser stays signed in between jobs), send and receive email from the computer's own address, fill in forms, download and edit files, and jobs that take longer than a chat reply or should keep going after the user leaves. Write the task the way you would brief a contractor: goal, details, and what to report back; include only what the task needs, not the conversation. Returns the agent's answer when it finishes, plus replay_url: a recording of the screen while it worked. Give the user that link so they can watch what their computer did. Big jobs can take several minutes. Uses the user's MyClawn plan.
Inputs: message (required), private. Returns: reply, status, replay_url, replay_note, +1 more
myclawn_desktop_send_fileGive the computer a fileAIdempotentInspect
Puts a text file on the MyClawn computer's desktop (/home/user/Desktop) so its agent can read or use all of it, for example a list of contacts, a draft, or data to enter. Use it instead of pasting long text into a task, then refer to the file by name in the task. A file with the same name is replaced. Text only, up to 2 MB. Returns the saved name and path; state is no_machine when the user has no computer yet.
Inputs: name (required), content (required). Returns: name, path, state
myclawn_desktop_sent_mailRead the email it sentARead-onlyIdempotentInspect
Lists every email the MyClawn computer sent from its own address, read from the mail server's copy (recipient, subject, time). Use it to confirm a message really went out, or to show the user what was sent. Read-only. state is ok, no_machine (no computer yet) or unavailable (the mail copy could not be read just now).
Inputs: no named parameters. Returns: state, messages
myclawn_desktop_statusMyClawn computer statusARead-onlyIdempotentInspect
Shows the user's MyClawn plan, any free trial and its hours left, whether their computer is ready, and the computer's own email address. Use it when the user asks about their MyClawn setup, or after a task said the computer is still starting. Read-only; fields are null when there is no plan or computer yet.
Inputs: no named parameters. Returns: plan, trial, machine, agent_email, +1 more
myclawn_desktop_tasksList past tasksARead-onlyIdempotentInspect
Lists the tasks the MyClawn computer has run, newest first, with each task's title, when it last changed and its state (done, or still running). Use it when the user asks what the computer did earlier or whether a job finished. Read-only; returns an empty list when the computer is not set up or not running.
Inputs: no named parameters. Returns: tasks
myclawn_notes_recallLook up notesARead-onlyIdempotentInspect
Searches the notes saved in the user's MyClawn memory by keyword and returns up to 5 that share a word with the search, best match first (an empty list when nothing matches). Use it when the user refers to something they asked you to remember. Read-only.
Inputs: query (required). Returns: notes
myclawn_notes_rememberRemember a noteAInspect
Save a note to the user's MyClawn memory, which their MyClawn computer and every assistant connected to it can use. Only when the user asks you to keep something ("remember that…"). Never save conversation content on your own.
Inputs: text (required), kind. Returns: saved
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Persistent Linux computers for AI agents: desktop, signed-in browser, vault, apps and shell.
Cloud desktops for AI agents: create a Linux computer, see its screen, click, type and run tasks.
Your own browser in the cloud, logged in to your sites, shared by you and your AI.
Persistent cloud workspaces for AI agents: run commands, edit files, use git and a browser.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables AI coding agents to rent disposable cloud dev environments with a terminal, real browser, repo checkout, git gateway, and live view, so they can build, test, run dev servers, inspect pages, and push branches while keeping keys off the VM.1Apache 2.0
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to operate a real, private, local Docker-based computer with durable files, terminal access, web research, and a persistent browser or desktop. It supports multiple agent clients via MCP or OpenAPI, with live viewing and human takeover.13Apache 2.0
- AlicenseNot gradedqualityBmaintenanceLets coding agents drive a real headful Chromium over CDP on a remote headless box, giving each project its own persistent profile, isolated login sessions, downloads, and on-demand browser lifecycle. When a site needs a human for sign-in, a captcha, 2FA, or a passkey, the operator opens the same browser from their own machine, completes the step, and hands control back to the agent.Functional Source , Version 1.1, ALv2 Future
- FlicenseNot gradedqualityDmaintenanceEnables autonomous desktop automation by delegating tasks to vision-based agents operating within cloud-based virtual machine sandboxes. It allows users to manage VMs, execute complex computer tasks, and receive text-based screen summaries across Linux, Windows, and macOS environments.2-
Glama MCP Gateway
Add one secure layer between your agents and this server.