Borrowing it
Nothing to install: this file belongs to ju-li/automata-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ju-li/automata-mcp/main/CLAUDE.mdgit clone --depth 1 https://github.com/ju-li/automata-mcpWrote 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/ju-li/automata-mcp/claude-md)<a href="https://agentmods.dev/instructions/ju-li/automata-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/ju-li/automata-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/ju-li/automata-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/ju-li/automata-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.26755 | $0.26755 |
| Opus 5.5 | $0.10702 | $0.10702 |
| Sonnet 5.5 | $0.05351 | $0.05351 |
| Haiku 4.5 | $0.02676 | $0.02676 |
Grade C, and why
automata-mcp CLAUDE.md scanned grade C with 2 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 today.
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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **Evolution migrations run as root** (`evolution-migrate`, upstream script as child — not `set -u` clean, so not sourced); everything else drops to `node` / `postgres` via `s6-setuidgid`. Evolution's cwd-relative `inst Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
- **Directory listings repeat version: `server.json` (official MCP Registry), `lhm.plugin.json` (LobeHub), `.cursor-plugin/plugin.json` (cursor.directory, beside root `mcp.json`).** `.github/scripts/listings.mjs` = one r How it starts
The opening of the file, as written. The whole thing — 671 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code (claude.ai/code) working in this repo.
What this is
One Nuxt/Nitro server, two surfaces:
- Web UI — users create connections, mint connector tokens for them. Session cookie.
- MCP endpoint — Claude connects to
/mcpas custom connector. App-minted bearer token.
Connection = row in instances with kind; connector token bound to exactly one row. Four kinds:
whatsapp— Evolution API instance, paired by QR, on deployment's configured Evolution server or user-supplied one.postgres— user-supplied PostgreSQL DSN.telegram— personal Telegram account linked by QR throughapps/telegram-bridge, deployment's own bridge or user-supplied one. See "Telegram bridge".imap— user-supplied IMAP mailbox, username + app password. Read-only except saving drafts. Headers synced, bodies and attachments fetched live. See "IMAP connections".
kind read through instanceKind() in mcp-scope.ts. Column NOT NULL with CHECK, but function still refuses value running build doesn't know rather than serving it as WhatsApp — rollback after migration added a kind leaves such rows. Every per-kind branch = switch ending in assertNever (shared/connection.ts), never if … else, so adding kind fails to compile everywhere not yet decided.
App's own data (users, sessions, organizations, connections + credentials, tokens) in Postgres, app schema migrated at boot — NUXT_DATABASE_URL, required. Same Postgres Evolution writes into (public) and Telegram bridge uses (telegram); that Postgres is only compose service not behind profile. Evolution API + its Redis needed only for WhatsApp connections — behind whatsapp compose profile; NUXT_EVOLUTION_* vars optional.
README.md = public landing page (logo, links, screenshots, features, quick start) — keep short. Operator's manual lives in docs/: self-hosting.md (root docker-compose.yml + .env.selfhost.example, published images), deploy-railway.md, development.md (local setup, networking tables, Linux firewall rule), whatsapp.md, telegram.md, postgres.md, imap.md, alerts.md, using-automata.md, releases.md, billing.md. Read before anything involving Docker or local stack; this file covers code. Two compose files: docker-compose.dev.yml = services only for dev; docker-compose.yml = full self-host stack, separate project name automata. Third self-host path: docker/aio/ = all-in-one image, everything in one container — see "All-in-one image".
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.
- today Changed · +2 lines · +513 tokens per session scan B → C de517a7720c5
- 5d ago Changed · +37 lines · +1,622 tokens per session 6ce6b7377684
- 8d ago Changed · +28 lines · +1,014 tokens per session c47671bb7731
- 9d ago Changed · +40 lines · +1,736 tokens per session b08d8390f319
- 12d ago Changed · +17 lines · +897 tokens per session scan A → B 6d29822e0649
- 17d ago Changed · -35 lines · -3,294 tokens per session 1ff5b0b0b7a8
- 24d ago Changed · +19 lines · +884 tokens per session 6b7dabbf6a65
- 29d ago Changed · +60 tokens per session 3efcdacb8c06
- 1mo ago Changed · +4 lines · +523 tokens per session 2703e41b2c66
- 1mo ago First seen · 559 lines · 22,800 tokens per session scan A 259c4a231f12
automata-mcp CLAUDE.md is an instructions file published in the GitHub repository ju-li/automata-mcp (6 stars, last pushed today), licensed MIT. It adds 26,755 tokens to every session, about $0.1070 per session on Opus 5.5. A static security scan graded it C with 2 findings (asks for root, reads mcp configuration). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.