Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/haliphax-ai/openclaw-tools-mcp-server/agents-mdgit clone --depth 1 https://github.com/haliphax-ai/openclaw-tools-mcp-serverWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/haliphax-ai/openclaw-tools-mcp-server/agents-md)<a href="https://agentmods.dev/instructions/haliphax-ai/openclaw-tools-mcp-server/agents-md"><img src="https://agentmods.dev/badge/instructions/haliphax-ai/openclaw-tools-mcp-server/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00814 | $0.00814 |
| Opus 5 | $0.00407 | $0.00407 |
| Sonnet 5 | $0.00163 | $0.00163 |
| Haiku 4.5 | $0.00081 | $0.00081 |
Grade A, and why
openclaw-tools-mcp-server AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — openclaw-tools-mcp
Instructions for AI agents working on this codebase.
What this is
An MCP (Model Context Protocol) server that proxies OpenClaw tool calls from ACP agents to the OpenClaw Gateway's HTTP API. It runs as a stdio child process spawned by the ACP agent.
Project structure
src/
index.ts Entry point — creates McpServer, registers tools, connects stdio
config.ts Configuration loading (env vars + optional JSON config file)
gateway.ts HTTP client for POST /tools/invoke on the OpenClaw Gateway
custom-tools.ts Loads additional tool definitions from JSON files on disk
tools/
index.ts Tool registry — registers enabled tools on the McpServer
web-search.ts web_search tool schema
web-fetch.ts web_fetch tool schema
browser.ts browser tool schema (action-based, many sub-actions)
pdf.ts pdf tool schema
message.ts message tool schema (action-based)
canvas.ts canvas tool schema (action-based)
sessions-send.ts sessions_send tool schema (agent-to-agent messaging)
tools/
rooms.json Custom tool definition for the rooms plugin (5 tools)
How it works
- ACP agent spawns the server via stdio
- Agent sends MCP
tools/callrequests - Server proxies each call to
POST http://<gateway>/tools/invokewith bearer auth - Gateway executes the tool and returns the result
- Server wraps the result and returns it via MCP JSON-RPC
Key patterns
- Built-in tools are defined in
src/tools/*.tsas Zod schemas. Each exports aToolDefwithname,description,schema. - Custom tools are loaded from JSON files at paths specified in config. They use JSON Schema format (same as OpenClaw's
registerTool()). - sessionKey handling: Most tools get an auto-injected
sessionKeyparam for gateway routing (stripped before forwarding asargs). Tools that define their ownsessionKeyin their schema (likesessions_send) keep it as a tool argument — it's not stripped. - Result unwrapping: Gateway results may be wrapped in
{ content, details }. The handler extractsdetailswhen present for cleaner output.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 92 lines · 814 tokens per session scan A 3c4b8b04e659
openclaw-tools-mcp-server AGENTS.md is an instructions file published in the GitHub repository haliphax-ai/openclaw-tools-mcp-server (0 stars, last pushed 5mo ago), licensed Unlicense. It adds 814 tokens to every session, about $0.0041 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
zeroclaw AGENTS.md
AGENTS.md instructions for zeroclaw-labs/zeroclaw, covering agents.md - zeroclaw, single source of truth, safety and privacy, working rules and user-facing text.
zeroclaw CLAUDE.md
Claude Code instructions for zeroclaw-labs/zeroclaw, covering claude.md — zeroclaw (claude code), claude code settings, hooks and slash commands.
jentic-one CLAUDE.md
Claude Code instructions for jentic/jentic-one, covering claude.md, quick start, rules & guides, commands and project layout.
paiml-mcp-agent-toolkit CLAUDE.md
Claude Code instructions for paiml/paiml-mcp-agent-toolkit, covering claude code configuration, batuta stack (sovereign ai tools), dependencies requiring review, critical: code search policy and search mode flags.
wanaku CLAUDE.md
Claude Code instructions for wanaku-ai/wanaku, covering wanaku, core guidelines, quick start, build info and architecture.
paiml-mcp-agent-toolkit AGENTS.md
AGENTS.md instructions for paiml/paiml-mcp-agent-toolkit, covering antigravity global rules for pmat, agent guidelines and contract-first, and how it is enforced.