25 thinking prompts distilled into one agent-routing skill 'laoyeye'
卡尔的AI沃茨 · wechat · 2026-10-20
The author, inspired by Matt Pocock's ask-matt routing skill, merged 25 daily-use thinking prompts (Socratic questioning, first principles, double steelman, Anthropic's allegory memory method, Johari window, inversion, etc.) into one routing skill called 'laoyeye', organized into six scenarios: clarifying problems, learning, solving, deciding, planning, and self-knowledge.
- Problem solved: you save many prompts but forget to use them at the right moment; a router lets you describe your situation and have the AI pick the right prompt, explain why not another one, and what to chain next.
- Install: npx skills@latest add LearnPrompt/laoyeye, then say 'laoyeye 帮我' anytime (works great with voice input).
- Key craft rule: never rewrite the original prompts — compressing Socratic questioning into 'ask step by step' made the model dump all six questions at once, destroying the prompt's value.
- Deep-dives two methods: Johari window (AI maps what you know, probes blind spots, then stitches both) and inversion (list all failure points, flip each into a must-follow checklist).
- Names are prefixed (matt-grilling, etc.) so nothing conflicts with the upstream skills repo.
More from coding & agent
- Claude Code Pushes Urgent Security Update Across All Installations — umeshonai · 2026-10-23
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23