openclaw-tools-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OPENCLAW_MCP_TOOLS | No | Comma-separated list of tools to expose (default: all 7 tools) | |
| OPENCLAW_MCP_CONFIG | No | Path to config file | openclaw-mcp.json |
| OPENCLAW_GATEWAY_URL | No | Gateway base URL | http://127.0.0.1:18789 |
| OPENCLAW_GATEWAY_TOKEN | Yes | Bearer token for gateway auth | |
| OPENCLAW_MCP_CUSTOM_TOOLS | No | Comma-separated paths to custom tool JSON files |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| web_searchC | Search the web and return results |
| web_fetchA | Fetch and extract readable content from a URL (HTML converted to markdown or text) |
| browserC | Control a web browser. Actions: status, start, stop, tabs, open, snapshot, screenshot, navigate, act, click, type, scroll, select, hover, wait, evaluate, close, pdf |
| pdfB | Analyze PDF documents from file paths or URLs |
| messageC | Send messages and perform channel actions (Discord, Telegram, WhatsApp, etc.). Actions: send, react, search, read, poll, thread-create, thread-reply |
| canvasC | Control the Canvas visual workspace. Actions: present, hide, navigate, eval, snapshot, a2ui_push, a2ui_reset |
| sessions_sendA | Send a message into another OpenClaw session. Use for agent-to-agent communication. Either sessionKey or label must be provided to identify the target session. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Each tool targets a distinct domain: web search, URL fetching, browser control, PDF analysis, messaging, canvas, and inter-session communication. No two tools have overlapping purposes; even web_fetch and browser are differentiated by control vs extraction.
The naming pattern is inconsistent: some tools use verb_noun (web_search, web_fetch), some are single nouns (browser, pdf, message, canvas), and one uses noun_verb (sessions_send). All are lowercase snake_case, but the structure varies.
Seven tools is well-scoped for a general-purpose assistant toolkit, covering major capabilities like web, communication, document analysis, and workspace control without feeling bloated or sparse.
The suite covers web research, browsing, document analysis, messaging, canvas control, and agent-to-agent communication. Minor gaps like calendar or file management could be expected, but the core workflows are well covered.