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/archolith/menhir/agents-mdgit clone --depth 1 https://github.com/Archolith/menhirWhat 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 | $0.00395 | $0.00395 |
| Opus 5 | $0.00198 | $0.00198 |
| Sonnet 5 | $0.00079 | $0.00079 |
| Haiku 4.5 | $0.00040 | $0.00040 |
Grade A, and why
menhir 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.
What it actually says
Menhir agent instructions
These instructions apply to the entire repository.
Start here
- Read
.agent/README.md. It is a router; do not preload all of.agent/. - Follow the task-specific route it names. Read large architecture, data-model, endpoint, or memory-design references only when the task requires them.
- Before code exploration, confirm Menhir appears in
query_structure(query_type="projects"). Use structural queries before filesystem search, and checkblast_radiusonce per file before editing. - Verify the repository root, current branch, and worktree status. Preserve unrelated changes and stage files explicitly.
Implementation rules
- Python 3.12+, builtin generics, explicit public return types, and small composable modules.
- Keep external LLM, embedding, Neo4j, and auth configuration in environment variables. Never commit or record secrets.
- Extend the canonical backend/runtime and MCP contracts instead of creating parallel paths.
- Treat stale structural anchors and incomplete indexes as inconclusive until current code is checked.
- Use
.agent/workflows/feature_planning.mdbefore a semi-large or cross-cutting change. - Use
.agent/workflows/code_conventions.mdfor code style and.agent/maintenance.mdfor changelog and closeout requirements.
Verification
- Prefer focused tests from
query_structure(query_type="affected_tests"). - Routine safe suite:
pytest tests/ -m unit -q(serial; do not use-n autoon the maintainer machine). - Live Neo4j/LLM tests require explicit opt-in and the documented environment.
- Report commands actually run and distinguish failures from tests that were not run.
For instructions that consumer agents can copy into repositories using Menhir, see
docs/templates/AGENTS.menhir.md and docs/agent-usage.md.
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 First seen · 35 lines · 395 tokens per session scan A 39ca9cbca274
menhir AGENTS.md is an instructions file published in the GitHub repository Archolith/menhir (2 stars, last pushed 3d ago), licensed Apache-2.0. It adds 395 tokens to every session, about $0.0020 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
remembrallmcp CLAUDE.md
Claude Code instructions for roboticforce/remembrallmcp, covering remembrallmcp, quick reference, build & run, start database and build everything.
remembrallmcp AGENTS.md
AGENTS.md instructions for roboticforce/remembrallmcp, covering remembrallmcp, scope, project overview, common commands and cli reference.
repobrain copilot-instructions.md
Copilot instructions for study8677/repobrain, covering github copilot bootstrap instructions and hard rule — query the repobrain hub first.
repobrain AGENTS.md
AGENTS.md instructions for study8677/repobrain: Beautiful is better than ugly. Explicit is better than implicit. Simple is better than complex. Complex is better than complicated. Flat is better than nested. Sparse is better than dense. Readability counts. Special cases aren't special enough to break the rules.…
Ctxo CLAUDE.md
Instructions for alperhankendi/Ctxo, covering claude.md — ctxo, project overview, quick reference, dev (pnpm workspace) and usage (consumer).
trace-mcp AGENTS.md
Instructions for nikolai-vysotskyi/trace-mcp: IMPORTANT: For ANY code exploration task, ALWAYS use trace-mcp tools first. NEVER use Read/Grep/Glob/Bash(ls,find) for navigating source code.