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/with-logic/crew/claude-mdgit clone --depth 1 https://github.com/with-logic/crewWhat 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 | $0.07161 | $0.07161 |
| Opus 5 | $0.03580 | $0.03580 |
| Sonnet 5 | $0.01432 | $0.01432 |
| Haiku 4.5 | $0.00716 | $0.00716 |
Grade C, and why
crew 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 3d 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.
doesn't regenerate cleanly. Prefer `rm -rf coverage/ && bun test` to Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **Don't mock `git`** at the `Bun.spawnSync` level. Use the How it starts
The opening of the file, as written. The whole thing — 629 lines — stays where its author put it; the contents beside it link to each section on GitHub.
crew — Contributor Guide
This file is the briefing for anyone (human or AI) working on crew.
Read it before making changes.
crew is a package manager for Agent Skills specified by PRD.md.
The PRD is the contract; this document is about how we implement it.
Working agreement: specification changes
If you are asked to make a change that is not specific to our
implementation — that is, a change that any conformant implementation
of the spec would also need to make — update PRD.md first.
Examples of what triggers a PRD update:
- A new command, flag, or subcommand.
- A new error name or a change to an exit code.
- A change to the marker format, state schema, or config schema.
- A change to the reference grammar or resolution precedence.
- A change to a default (tap URL, autoupdate interval, lock timeout, …).
- A change to the hash algorithm or what it covers.
- Adding, removing, or renaming an agent adapter (the set in §7.2).
- A change that shifts any of the §18 conformance criteria.
Examples of what does not need a PRD update:
- Internal refactors, file moves, renamed internal functions.
- Test additions or helper tweaks.
- Dependency bumps that don't change behavior.
- Bug fixes that make behavior match the existing spec.
- Performance improvements within existing guarantees.
When in doubt: if an external observer running the conformance suite could notice the change, update the PRD. The PRD is what a second implementation would read to build the same tool — keep it accurate.
After updating the PRD, update src/ and tests/ to match. Don't
leave a PRD change unimplemented across a commit.
Working agreement: user-facing surfaces
If a change affects user-visible behavior — a new command, a new flag, a change to install/update/tap semantics, a new way for a tap to be structured, a renamed default, anything a user of the CLI would notice — also check whether these user-facing surfaces describe the thing you just changed, and keep them in sync:
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.
- 3d ago First seen · 629 lines · 7,161 tokens per session scan C abd1374efab0
crew CLAUDE.md is an instructions file published in the GitHub repository with-logic/crew (47 stars, last pushed 9d ago), licensed MIT. It adds 7,161 tokens to every session, about $0.0358 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
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.
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.
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 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).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.