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/luisa-sys/lyra-mcp-server/claude-mdgit clone --depth 1 https://github.com/luisa-sys/lyra-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/luisa-sys/lyra-mcp-server/claude-md)<a href="https://agentmods.dev/instructions/luisa-sys/lyra-mcp-server/claude-md"><img src="https://agentmods.dev/badge/instructions/luisa-sys/lyra-mcp-server/claude-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 | $0.05576 | $0.05576 |
| Opus 5 | $0.02788 | $0.02788 |
| Sonnet 5 | $0.01115 | $0.01115 |
| Haiku 4.5 | $0.00558 | $0.00558 |
Grade C, and why
lyra-mcp-server 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 5d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
**Cleanup is mandatory at end of session:** every worktree Claude created must be either removed (work merged or abandoned) or explicitly noted as "kept for next session" in the session summary. Use `git worktree remove Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s https://mcp.checklyra.com/.well-known/mcp.json | jq -r .build_sha How it starts
The opening of the file, as written. The whole thing — 266 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — Project Instructions for Claude
This file contains instructions and policies that Claude must follow when working on this repository.
Pre-Work Checklist
Before starting any task, Claude must:
- Check Jira — confirm a ticket exists for the work, or create one. Never start work without a tracked ticket.
- Check the lyra repo's docs/ — architecture docs live in the main lyra repo at
docs/. Read relevant docs before acting on architecture, ops, or infrastructure questions. - Check for existing work — search the codebase and recent PRs to avoid duplicating effort.
- Run tests before and after — every change must leave tests green.
- Confirm working tree isolation — if any other Claude Code instance might be running against this repo, this session MUST run inside its own git worktree (see "Parallel Claude sessions" below). Verify with
git branch --show-currentat the start of work AND right before every commit; if HEAD has changed unexpectedly, stop and recover (BUGS-17).
Parallel Claude sessions — use git worktrees
Luisa runs multiple Claude Code instances in parallel to work on independent features. All instances after the first MUST operate in a git worktree rather than the shared main checkout. Without isolation, parallel processes switch HEAD mid-flow and contaminate each other's commits — see BUGS-17 for the canonical incident (KAN-69a work landed on top of KAN-191 commits because two Claude sessions were sharing one checkout).
The lyra repo's CLAUDE.md is the source of truth for this policy. Briefly:
- Use
EnterWorktree(Claude Code built-in) to isolate the current session. - For sub-agents, pass
isolation: "worktree"on the Agent tool call. - For manually-launched sessions,
git worktree add ../<repo>-<branch> origin/main(ororigin/developin the lyra repo) before invokingclaude. - Pre-commit safety check:
git branch --show-currentmust equal the branch you intended. If it doesn't, do not commit — recover per BUGS-17's recovery section.
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.
- 5d ago First seen · 266 lines · 5,576 tokens per session scan C 5e0bf97a606a
lyra-mcp-server CLAUDE.md is an instructions file published in the GitHub repository luisa-sys/lyra-mcp-server (0 stars, last pushed yesterday), licensed MIT. It adds 5,576 tokens to every session, about $0.0279 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). 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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.