// MCP TOOL REFERENCE

OPENCLAW TOOLS

9 tools from the OpenClaw MCP Server, categorised by risk level.

READ TOOLS

attachments_fetch Extract non-text content metadata from messages 2/5 conversation_get Retrieve a single conversation by session key 2/5 conversations_list List recent routed conversations with filtering 2/5 events_poll Retrieve queued live events since a cursor position 2/5 events_wait Long-poll for the next matching event with timeout 2/5 messages_read Read historical transcript messages from a conversation 2/5 permissions_list_open Show pending approval requests observed since connection 2/5

WRITE TOOLS

messages_send Send text replies through Slack, Discord, Telegram, and 50+ platforms 5/5 permissions_respond Resolve approvals with allow-once, allow-always, or deny 5/5

// FAQ

How many tools does the OpenClaw MCP server have? +

The OpenClaw MCP server exposes 9 tools across 2 categories: Read, Write.

How do I enforce policies on OpenClaw tools? +

Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the OpenClaw server.

What risk categories do OpenClaw tools fall into? +

OpenClaw tools are categorised as Read (7), Write (2). Each category has a recommended default policy.

Enforce policies on OpenClaw

Open source. One binary. Zero dependencies.

npx -y @policylayer/intercept

github.com/policylayer/intercept →