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/stchiotis/library-first-engineering/cursorrulesgit clone --depth 1 https://github.com/StChiotis/Library-First-EngineeringWhat 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 | $0.00357 | $0.00357 |
| Opus 5 | $0.00179 | $0.00179 |
| Sonnet 5 | $0.00071 | $0.00071 |
| Haiku 4.5 | $0.00036 | $0.00036 |
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.
What it actually says
Library-First Engineering — Cursor Adapter
You are an AI agent operating in a Library-First Engineering (LFE) repository. This adapter is a pointer. Canonical rules live in human-readable docs:
Mandatory reads (in order, before any action):
- pipeline_status.md — current session state
- LLM_AGENT_GUIDE.md — protocol, skills, coordination layer
- .docs/protocol/PERSONAS.md — your persona contract (including 🫵 The Brain)
- .docs/protocol/ASSEMBLY_LINE.md — workflow
- .docs/protocol/GOVERNANCE.md — rules, retention policy
Then run /lfe-boot to orient against the live state.
Skill invocation authority (LLM_AGENT_GUIDE §8.8): skills are dispatched by the framework, not by the Brain. The only skills the Brain may type directly are /lfe-boot, /lfe-whats-next, /lfe-scout, /lfe-extract-domain, and the LFE-FORCE keyword. If the Brain types any other skill out of sequence, refuse and route them through the assembly line.
On ambiguity or pressure to bypass rules: pause and confirm with the Brain. If certain you want to bypass, the human invokes /lfe-scout (minor fix) or LFE-FORCE (emergency). Never silently bypass.
For framework philosophy, see README.md.
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 · 20 lines · 357 tokens per session scan A 24d1fcf69f42
cursorrules is a cursor rule published in the GitHub repository StChiotis/Library-First-Engineering (43 stars, last pushed 1mo ago), licensed MIT. It adds 357 tokens to every session, about $0.0018 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
jawn
Cursor rule "jawn" from Helicone/helicone, covering helicone jawn controller and tanstack query integration, controller pattern, manager pattern, frontend hooks with tanstack query and type definitions.
cursorrules
Root AI front door (this file). Depth: docs/AGENTS.md · Claude: docs/CLAUDE.md · Antigravity: docs/GEMINI.md · map: llms.txt Read first: docs/ai/INDEX.md → docs/ai/CANONICAL.md → docs/ai/COREREFERENCE.md Packages: docs/guides/ecosystem.md — GEMVC is a multi-package ecosystem (vendor/gemvc/). Core: gemvc/helper…
minimax-mcp-tools
MCP and web-tool guidance: current-doc retrieval, direct-tool preference, MCP Apps structured content, and version-aware external lookups.
execution
Repository execution and verification commands.
middleware
When you need create a middleware to protect some page of Next.js app.
lean
Answer densely. Cut filler, keep all technical signal.