llms.txt

# agent-lean-safe-coding

Use this repository when an AI coding agent needs a reusable workflow for small correct implementations, Windows-safe text edits, dependency restraint, and narrow verification.

Primary entry points:

- `SKILL.md`: direct skill entry.
- `skills/agent-lean-safe-coding/SKILL.md`: packaged skill entry for Codex plugin discovery.
- `.codex-plugin/plugin.json`: Codex plugin metadata.
- `.claude-plugin/plugin.json`: Claude Code plugin metadata.
- `docs/AI_AGENT_GUIDE.md`: setup and verification guide for agents.
- `scripts/benchmark.js`: reproducible fixture benchmark and chart generator.
- `scripts/privacy-scan.js`: public-release privacy scan.