pi
Reduces cognitive overhead in AI prompt design by applying structured strategic principles, improving consistency and quality of outputs across teams without…
A prompt-discipline engine for AI coding agents that blends Chinese classical strategy (Art of War and the Hundred Schools), Zen practice, MBTI cognitive archetypes, and Western methodologies into a structured instruction system. Installs as a skill for Claude Code, Codex CLI, Cursor, Gemini CLI, and 10+ other agents, adding difficulty tiers (light/standard/deep), scenario routing, verification matrices, and forced task decomposition to agent behavior. Documentation is primarily Chinese with an English variant (pi-en).
- Add structured plan-verify discipline to Claude Code sessions via light/standard/deep difficulty tiers
- Route tasks to predefined scenarios (coding, testing, product, ops) with keyword-based scenario selection
- Enforce task decomposition on large changes — forced split above 3 files with per-step verification
Reduces cognitive overhead in AI prompt design by applying structured strategic principles, improving consistency and quality of outputs across teams without requiring deep ML expertise.
Development teams integrating AI models who need repeatable, documented prompt patterns and want to minimize trial-and-error in production systems.
https://github.com/share-skills/pi
By share-skills
How to Get It
curl -fsSL https://raw.githubusercontent.com/share-skills/pi/main/install.sh | bash
Tip: Paste this into a Claude Code conversation. Verify command matches your Claude Code version.
After installing, paste this into Claude:
Help me add structured plan-verify discipline to Claude Code sessions via light/standard/deep difficulty tiers
Trust Signals Auto-scanned
Reviewer notes
Auto-scanned review. These are observations, not a security certification.
Scored from trust signals (evidence-eval-v1): 100 GitHub stars; contributors unknown; last commit 2d ago; license Apache-2.0.
Things to check
- Documentation and much of the skill content are in Chinese; install the English variant (pi-en) if needed. The recommended install pipes a remote script to bash — review install.sh before running. Optional visualizer requires git and node/npm.
How to evaluate tools before deploying →
Data shown here comes from public APIs and automated scanning. Reviewer notes reflect one person's experience. This is not a security certification or legal recommendation. Always evaluate tools according to your own organization's policies.
Evaluation
Scored from trust signals (evidence-eval-v1): 100 GitHub stars; contributors unknown; last commit 2d ago; license Apache-2.0.