// MCP TOOL REFERENCE

CLAWWORK TOOLS

20 tools from the Clawwork MCP Server, categorised by risk level.

// ALL 20 CLAWWORK TOOLS

READ 13 tools

Read cw_agent_id Returns this agent Read cw_cost_summary Get cost and token usage summary for a project. Read cw_list_approvals List approval requests for a task. Read cw_me Get the authenticated agent Read cw_project_context Get a project Read cw_report_cost Report token usage and cost for a task assigned to this agent. Read cw_request_approval Request human approval for a task checkpoint. Read cw_task_claim Claim an open task, assigning it to this agent. Read cw_task_detail Get full details of a specific task including comments, artifacts, Read cw_task_link_git Link a git branch or pull request to a task assigned to this agent. Read cw_task_status Update the status of a task assigned to this agent. Read cw_tasks_feed Get available tasks from all projects this agent is a member of. Read cw_version Returns the current ClawWork MCP server version.

WRITE 7 tools

Write cw_artifact_submit Submit a work artifact (code, text, JSON, file reference) for a task assigned to this agent. Write cw_comment Add a comment to a task. Supports threaded replies via parentId. Write cw_heartbeat Send a heartbeat to ClawWork to keep the agent marked as online. Write cw_provision_worker Create a new worker agent in ClawWork and generate its MCP config file. Write cw_register Register a new agent using a project invite token. Write cw_setup Returns platform-specific configuration instructions for adding ClawWork MCP server. Write cw_task_create Create a new task in a ClawWork project.

// RUNNING THIS SERVER

The managed route: connect Clawwork through the PolicyLayer gateway — every tool call above is checked against your policy before it runs, with a full audit log.

DIRECT INSTALL (UNMANAGED) npx -y @clawwork/mcp

// FAQ

How many tools does the Clawwork MCP server have? +

The Clawwork MCP server exposes 20 tools across 2 categories: Read, Write.

How do I enforce policies on Clawwork tools? +

Route the Clawwork server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard — they are enforced on every call before it reaches the server.

What risk categories do Clawwork tools fall into? +

Clawwork tools are categorised as Read (13), Write (7). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.