agentleFS
kanekoshoyukanekoshoyu/hot-potato/skills/logical-thinking/SKILL.md

logical-thinking

Skill4 starsChanged 59 days agoMITView on GitHub

Use before any diagnosis — observe, verify, then solve.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. TheGoat395/Codex-Skills★ 136same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    怎么解决?: Only after 1–2. Smallest change that removes the *verified* cause.

    line 14
  2. Distinctive:

    区分「没证据」和「证据反对」。: No reply ≠ not delivered. delivered ≠ read ≠ acked — the state machine exists so you never conflate these.

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Logical Thinking — 观察 → 考证 → 解决

Sho's bar (2026-09-19): 观察和考证先行,解决方案是最简单的部分。这是习惯,不是天赋。每个诊断都走完这个环,无论多小。

The three questions (answer in order, in writing)

  1. 问题是什么? One sentence, no cause embedded. "prod pool's push to A's gateway returns 401" — not "A is ignoring us" (that is a conclusion, not a problem).
  2. 怎么确定?证据呢? List observations. Each must be reproducible (tool output, id, timestamp). Mark which hypothesis each observation supports or kills.
  3. 怎么解决? Only after 1–2. Smallest change that removes the verified cause.

Anti-story rules (the failure mode is always story-telling)

  • 解释必须吃下全部观察。 One counter-observation kills a hypothesis. A root cause that ignores half the data is a story. Live example (2026-09-19): "401 → she never got woken → no reply" ignored that the fleet twin was delivered seconds earlier — the contradiction was already in hand.
  • 一个数据点不构成因果链。 Name the discriminating experiment that separates remaining hypotheses BEFORE running it. The 401 explained the prod-side failure but not the silence; the silence needed its own test.
  • 先读手头数据,再发新探针。 Evidence collected but not consumed is the #1 embarrassment: the peer's gateway URL was sitting in a registry dump printed minutes before "discovering" it with a fresh probe.
  • 区分「没证据」和「证据反对」。 No reply ≠ not delivered. delivered ≠ read ≠ acked — the state machine exists so you never conflate these.

Context discipline (clones forget; write it down)

  • Facts bought with real probing (endpoints, auth models, who runs where, per-pool credentials) go to memory/skill the same turn. Unrecorded = lost at next clone spin-up; "I knew that" is worthless if it is not stored.
  • Scope facts with coordinates: which pool, which machine, which direction. "It works" without coordinates is noise.

Closing the loop

  • Fix, then re-verify the original observation, not a proxy.
  • Land the lesson where the next agent will trip: README for product behavior, skill for method, memory for environment facts.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.