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 rules/ousid/lean/leangit clone --depth 1 https://github.com/ousid/leanWhat 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 | $0.00221 | $0.00221 |
| Opus 5 | $0.00111 | $0.00111 |
| Sonnet 5 | $0.00044 | $0.00044 |
| Haiku 4.5 | $0.00022 | $0.00022 |
Grade A, and why
lean 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.
What it actually says
Answer densely. Cut filler, keep all technical signal.
- Start with the answer. No "Sure", "I'd be happy to", "Let me", no restating the question.
- Fragments over sentences, one idea per line, no closing summary.
- NEVER change code, commands, paths, error strings, config values, or identifiers. Keep them byte for byte exact. Compress prose only.
- GATE: compress only when the reply is mostly prose AND longer than ~40 words. Short replies, pure code blocks or diffs, and tool heavy edits get answered normally with no scaffolding.
- Laravel/PHP: keep framework terms verbatim (Eloquent, Artisan, Blade,
php artisan, facades). Assume framework fluency, do not re-explain basics. For a bug, name theClass::method, then the cause, then the fix.
Do not print token savings figures. Measurement is external (lean-stats).
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 · 15 lines · 221 tokens per session scan A c996a9dd1c18
lean is a cursor rule published in the GitHub repository ousid/lean (4 stars, last pushed 1mo ago), licensed MIT. It adds 221 tokens to every session, about $0.0011 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 cursor rules, from other repositories
16-wechat-sdk
微信 SDK 接入规范（EasyWeChat / WechatManager）.
15-config-center
动态配置中心读取规范（ConfigCenter / configcenter）.
00-project
基于 Laravel + Blade + Vue3 + Element Plus 的轻量级后台管理系统.
99-guard
Cursor rule "99-guard" from neoshopcn/neo-admin, covering 🔒 protected directories（禁止修改）, ⚠️ conditional（仅在明确要求时允许） and 🚫 dangerous operations（禁止主动执行）.
20-frontend
Vue3 + Element Plus frontend rules.
10-backend
Laravel Admin 后端开发规范.