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/developyn/laver-mcp/agents-mdgit clone --depth 1 https://github.com/Developyn/laver-mcpWhat 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 | $0.00574 | $0.00574 |
| Opus 5 | $0.00287 | $0.00287 |
| Sonnet 5 | $0.00115 | $0.00115 |
| Haiku 4.5 | $0.00057 | $0.00057 |
Grade A, and why
laver-mcp 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 yesterday.
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 — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — @laver/mcp
MCP server for Laver. Every tool is a thin call to the public REST API at https://api.laver.app; there is no local state and no second implementation of anything. See README.md for the tool list and setup.
This repository mirrors what is published to npm as @laver/mcp. server.js, README.md and LICENSE are the published files (see files in package.json); anything else here is repo-only and does not ship.
Development
Source of truth for the server is the private Laver monorepo at mcp/. Changes land there, are published to npm, and are mirrored here — do not fix a bug only in this repo, or the next publish reverts it.
| Command | Purpose |
|---|---|
npm run check |
Full check suite against a stub API |
npm run check:live |
Same suite against the live API (needs LAVER_API_KEY) |
npm start |
Run the server on stdio |
Node 24 is the floor in engines.
The Dockerfile is repo-only and is not published to npm. It exists because directory listings (Glama) build the server and probe it. Build and probe it the same way they do:
docker build -t laver-mcp .
printf '%s\n' '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"probe","version":"1"}}}' \
'{"jsonrpc":"2.0","method":"notifications/initialized"}' \
'{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}' | docker run -i --rm laver-mcp
That must return all tools without LAVER_API_KEY set — the key is only needed once a tool actually calls the API. A server that refused to start without credentials would fail the listing checks.
Git rules
- Commit only when explicitly asked, and stage only intended files.
- No AI attribution anywhere in the repo's public record: no
Co-Authored-Bytrailer naming an assistant, no "Generated with" footer in commit messages, PR titles or PR bodies, and no assistant named as an author. The commit is authored by the person who asked for it. - Naming Claude, Cursor or any other client in documentation is unrelated to the above and stays — this server supports them and the README says so.
- Never amend, force-push or reset hard unless explicitly requested.
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.
- yesterday First seen · 36 lines · 574 tokens per session scan A e28d41f9310e
laver-mcp AGENTS.md is an instructions file published in the GitHub repository Developyn/laver-mcp (0 stars, last pushed 6d ago), licensed MIT. It adds 574 tokens to every session, about $0.0029 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
roadmap-skill AGENTS.md
Instructions for shiquda/roadmap-skill, covering project guidelines, git commit messages, format, types and examples.
flux AGENTS.md
Instructions for sirsjg/flux, covering agents.md, project overview, dogfooding, common commands and development.
operator CLAUDE.md
Instructions for untra/operator, covering claude.md - operator, project overview, tech stack, code style and plans & specs location.
orbit CLAUDE.md
Instructions for Noveum/orbit, covering orbit, hard rules, bun is the toolchain, not the runtime, layout and commands.
batty CLAUDE.md
Instructions for battysh/batty, covering batty — agent instructions, what is this project, tech stack, project structure and development principles.
FlowBoard AGENTS.md
Instructions for rasimme/FlowBoard, covering flowboard — agent entry point, start here and where things live.