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/rich627/whatsapp-claude-plugin/agents-mdgit clone --depth 1 https://github.com/Rich627/whatsapp-claude-pluginWrote 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/rich627/whatsapp-claude-plugin/agents-md)<a href="https://agentmods.dev/instructions/rich627/whatsapp-claude-plugin/agents-md"><img src="https://agentmods.dev/badge/instructions/rich627/whatsapp-claude-plugin/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 | $0.01346 | $0.01346 |
| Opus 5 | $0.00673 | $0.00673 |
| Sonnet 5 | $0.00269 | $0.00269 |
| Haiku 4.5 | $0.00135 | $0.00135 |
Grade A, and why
whatsapp-claude-plugin 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 2d 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI coding agents working in this repository. Read this file fully, then use the routing table at the bottom before starting any non-trivial task.
Project Overview
WhatsApp MCP Server plugin for Claude Code — connects WhatsApp as a messaging channel via the linked-device protocol (Baileys). Bidirectional messaging, media, voice transcription, access control, per-group AI personalities, cron tasks. Published on claude.com/plugins.
Tech Stack & Commands
- Runtime: Bun — TypeScript runs directly. No build step. No test suite.
- Deps (only 2 — no new dependencies without the user's explicit approval):
@modelcontextprotocol/sdk,@whiskeysockets/[email protected](4 known rc.9 bugs are patched bypatch-baileys.mjsvia postinstall). - Linting: Trunk (prettier, markdownlint, shellcheck, shfmt, checkov, trufflehog).
bun install # install deps (postinstall runs patch-baileys.mjs)
bun server.ts # run the MCP server
trunk check # lint
trunk fmt # format
Architecture
WhatsApp (phone) ←─ Baileys ─→ MCP Server (server.ts) ←─ stdio ─→ Claude Code
↓
~/.whatsapp-channel/ (runtime state, never in repo)
├─ access.json
├─ .baileys_auth/
├─ groups/<groupJid>/ (config.md = personality + cron,
│ memory.md = conversation memory)
├─ tasks.md (agent-maintained open-task list, read by catch_up)
└─ inbox/
server.ts— the entire MCP server in one file (~1900 lines; re-check withwc -lrather than trusting this number). MCP tools exposed to Claude:reply,react,download_attachment,edit_message,status,unreplied,catch_up,list_groups. Also contains: access-control engine (DM policiespairing/allowlist/disabled, group policies, pairing codes with 1-hour TTL, LID↔phone mapping), message pipeline (access gate → routing → mention detection → text/media extraction → optional transcription → 4096-char chunking inlengthornewlinemode peraccess.json'schunkMode), per-group personality loading, cron parser (reads the## Cron Jobssection — exactly that heading — in each group'sconfig.md).skills/— user-facing commands:/whatsapp-channel:setup,:configure,:access.hooks/hooks.json→hooks-handlers/session-start.sh— onboarding detection at session start.scripts/watchdog.sh,scripts/whisper-transcribe.sh— reference scripts users copy out of the repo (watchdog to~/.whatsapp-channel/watchdog.sh; the whisper script to~/whisper-transcribe.sh— server.ts hardcodes that home-dir path). Editing them in the repo does NOT affect running deployments until re-copied.
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.
- 2d ago Changed · +9 tokens per session bd5b3d35e615
- 6d ago First seen · 98 lines · 1,337 tokens per session scan A 14fe6f5abfac
whatsapp-claude-plugin AGENTS.md is an instructions file published in the GitHub repository Rich627/whatsapp-claude-plugin (79 stars, last pushed 3d ago), licensed Apache-2.0. It adds 1,346 tokens to every session, about $0.0067 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 2026-08-30.
Other instructions, from other repositories
openagent CLAUDE.md
Claude Code instructions for the-open-agent/openagent, covering claude.md, commands, architecture, backend (go / beego) and frontend (react).
gptme AGENTS.md
AGENTS.md instructions for gptme/gptme, covering agent instructions for gptme, git workflow, code style, testing and project structure.
AstrBot copilot-instructions.md
Copilot instructions for AstrBotDevs/AstrBot, covering astrbot development instructions, working effectively, bootstrap and install dependencies, running the application and dashboard build (vue.js/node.js).
zhin zhin-plugin.instructions.md
Instructions for zhinjs/zhin, covering zhin plugin runtime authoring, package contract, convention directories, imports and native typescript and command routes.
lobu AGENTS.md
AGENTS.md instructions for lobu-ai/lobu, covering repo map, unrecoverable — never do these, facts you cannot derive, ship a change and how to work.
HARTOS CLAUDE.md
Claude Code instructions for hertz-ai/HARTOS, covering claude.md, project overview, master key - ai exclusion zone, branch discipline — main branch only (mandatory) and sibling repos — canonical filesystem paths.