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/ihabkhaled/clawai/cursorrulesgit clone --depth 1 https://github.com/ihabkhaled/ClawAIWhat 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 | $0.00708 | $0.00708 |
| Opus 5 | $0.00354 | $0.00354 |
| Sonnet 5 | $0.00142 | $0.00142 |
| Haiku 4.5 | $0.00071 | $0.00071 |
Grade A, and why
cursorrules 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.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ClawAI — Cursor rules (compact router)
Do NOT invent repository facts. Before editing, run: npm run knowledge:context -- --task="" and read .ai/local/current-context.md.
Canonical authority (higher wins): CLAUDE.md > rules/00-non-negotiable-rules.md > context/architecture-map.md > context/stack-and-toolchain.md > numbered rules/* > skills/* > context/* + memory/* > .ai/manifests/*.
Blockers (never): --no-verify / hook bypass; eslint-disable / @ts-ignore / any / as unknown as; cross-service DB access; logic in controllers; DB calls outside repositories; process.env outside AppConfig; console.log; logging/exposing secrets; user-facing text without i18n (13 locales); code without a test; shipping a change with no knowledge delta (docs/skills/rules/context in the SAME commit, rules/33); gating per-commit or all-workspace instead of once at the end, scoped (rules/34).
Validation (touched folders only): npm run affected:list cd && npm run typecheck && npm run lint && npm test && npm run build
Scoped rules live in .cursor/rules/*.mdc (by file glob). Full router: cursor.md + AGENTS.md.
Prompt packs / execution prompts — before ANY code, run the protocol in rules/26-prompt-pack-intake-protocol.md (runbook: skills/execute-prompt-pack.md): read the pack end to end; npm run knowledge:context; read governing docs in authority order; audit every deliverable done/partial/missing against the CODE (present is not wired — a repo method with no callers is scaffolding); review the constraint surface up front (eslint flat config, TS strict, prettier, coverage floors, security, i18n x9 + i18n.types.ts, the CLAUDE.md delivery checklist, and the gate topology incl. pre-commit/pre-push/CI/freshness); write the plan and state deviations — policy outranks the pack, never silently; then implement with scoped gates, one gated commit per change, each pushed before the next.
Communication style (MANDATORY)
Short. Plain. Concrete. A few lines max.
- Blocked? One line:
Blocked: <the actual thing>. - Working?
Working — <what>.Progress?~70/100. - Name the concrete cause: file, symbol, exact error. Never circle the problem.
- Easy words over complex ones. Cut every reply in half before sending.
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 · 59 lines · 708 tokens per session scan A 987a5d4fa7a9
cursorrules is a cursor rule published in the GitHub repository ihabkhaled/ClawAI (22 stars, last pushed yesterday), licensed Apache-2.0. It adds 708 tokens to every session, about $0.0035 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 2026-08-30.
Other cursor rules, from other repositories
titanclip-monorepo
TitanClip pnpm monorepo layout, scripts, and scoping.
titanclip-cursor-adapter
Cursor local adapter — CLI integration and skills sync.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
typescript
Changes to these high-fan-out internals can affect every message, delta, element, or rerun. Keep work in them minimal, and benchmark changes with representative stress-test apps.
coolify-ai-docs
Master reference to all Coolify AI documentation in .ai/ directory.