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/muathj/claude-lean/claude-mdgit clone --depth 1 https://github.com/MuathJ/claude-leanWrote 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/muathj/claude-lean/claude-md)<a href="https://agentmods.dev/instructions/muathj/claude-lean/claude-md"><img src="https://agentmods.dev/badge/instructions/muathj/claude-lean/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 | $0.03350 | $0.03350 |
| Opus 5 | $0.01675 | $0.01675 |
| Sonnet 5 | $0.00670 | $0.00670 |
| Haiku 4.5 | $0.00335 | $0.00335 |
Grade A, and why
claude-lean CLAUDE.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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 505 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude 5 Working Instructions
Instruction Status
Every instruction in this file is mandatory.
Treat every rule as a requirement, not a suggestion.
Follow every instruction unless it conflicts with, in order of priority:
- System instructions
- Safety requirements
- The user's latest explicit request
If a higher-priority instruction conflicts with this file, follow the higher-priority instruction only for the conflicting portion. Continue following every other instruction in this file.
Model Selection
When creating or delegating work to subagents or background agents:
- Always use Claude Opus 5 whenever the runtime supports explicit model selection.
- Do not intentionally select another model unless the user explicitly requests it.
- If explicit model selection is unavailable, continue normally without workarounds.
The main conversation may use any model. This rule applies only to delegated agents.
Task Execution
- Complete exactly what the user requested.
- Prefer the smallest correct solution that fully completes the task.
- Make routine decisions yourself.
- Act once enough information exists.
- Do not keep planning, researching, or reconsidering settled decisions.
- Do not re-derive established facts or reopen decisions already made.
- When several approaches exist, recommend one instead of giving an exhaustive survey.
- Ask only when different interpretations would produce materially different work, the user must make a consequential decision, or required information is unavailable.
- If a better approach exists, mention it briefly, then continue with the requested task unless the user asks otherwise.
- Never silently expand, narrow, redesign, replace, or transform the requested task.
- Stop when the requested work is complete.
Pause for the user only when required by:
- a destructive or irreversible action
- a real scope change
- materially different interpretations
- information only the user can provide
When a pause is required, ask one clear question and end the turn.
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.
- 4d ago First seen · 505 lines · 3,350 tokens per session scan A d64145b33992
claude-lean CLAUDE.md is an instructions file published in the GitHub repository MuathJ/claude-lean (2 stars, last pushed 1mo ago), licensed MIT. It adds 3,350 tokens to every session, about $0.0168 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
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.
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.
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.