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/metagit-ai/metagit-cli/cursorrulesgit clone --depth 1 https://github.com/metagit-ai/metagit-cliWhat 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 | $0.00030 | $0.00387 |
| Opus 5 | $0.00015 | $0.00193 |
| Sonnet 5 | $0.00006 | $0.00077 |
| Haiku 4.5 | $0.00003 | $0.00039 |
Grade A, and why
agents 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
Metagit
After Every Task
After completing any task: update .mex/ROUTER.md project state and any .mex/ files that are now out of date. If no pattern existed for the task you just completed, create one in .mex/patterns/.
Before you tell the user the work is done — if this conversation introduced or edited project files under this repo — run task qa:prepush from the repository root (scripts/prepush-gate.py via scripts/prepush-gate.zsh). Fix failures and re-run until green. Only skip when the task was strictly read-only (no writes) or the user explicitly waived the QA gate in this thread.
Last step before hand-off — after QA passes (or when QA was skipped for read-only work), run task gitnexus:analyze from the repository root so the GitNexus index stays current for MCP tools and agents.
Commands (minimum)
- QA gate before hand-off / “work complete”:
task qa:prepush(format, lint, tests, integration tests, context-aware pip-audit/bandit; optionalgitleakswhen installed) - GitNexus index refresh (always last):
task gitnexus:analyze - Full unit tests only:
task test
Navigation
At the start of every session, read .mex/ROUTER.md before doing anything else.
For full project context, patterns, and task guidance — everything is there.
Use the GitNexus MCP tools to understand code, assess impact, and navigate safely.
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 · 25 lines · 30 tokens per session scan A bd1157af1a15
agents is a cursor rule published in the GitHub repository metagit-ai/metagit-cli (2 stars, last pushed 4d ago), licensed MIT. It adds 30 tokens to every session and 387 once invoked, about $0.0002 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
architector-onboarding
Onboard or sync a codebase into MCP Architector (phased, context-safe). Use when importing a new repo, documenting architecture, or refreshing architector data.
cuopt-skill-evolution
Trigger skill evolution — after solving non-trivial problems, propose skill updates.
cursorrules
See CLAUDE.md for full context. Quick rules.
moyu
当检测到过度工程模式时自动激活：修改用户未要求的代码、创建不必要的抽象层、 添加未要求的注释/文档/测试、引入不必要的依赖、重写整个文件而非最小编辑、 diff 超出请求范围。用户说"太多了"、"只改 X"、"简单点"时触发。 Activates on over-engineering: modifying unrequested code, unnecessary abstractions, unsolicited comments/docs/tests, unrequested dependencies, full file rewrites, scope-exceeding diffs. Triggers on "too…
cursorrules
This repository contains a structured skill library for AI-assisted product development. Skills are in skills//SKILL.md. Read the relevant SKILL.md before generating any artifact.
cursorrules
AgentAnycast is a decentralized P2P runtime for the A2A (Agent-to-Agent) protocol.