agentleFS
jshsakurajshsakura/awesome-opencode-skills/skills/first-principles-thinking/SKILL.md

first-principles-thinking

Skill30 starsChanged 4 months agoMITView on GitHub

Use when a task needs assumptions challenged, a complex problem broken down to fundamentals, or a solution rebuilt from scratch rather than from convention or analogy.

11 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 4

    Distinctive

    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:

    verify each assumption has an explicit verdict and supporting reasoning

    line 34
  2. Distinctive:

    ensure at least 2-3 rebuilt directions are generated before recommending one

    line 35
  3. Distinctive:

    check that recommendations trace back to fundamental truths, not convention

    line 36
  4. Distinctive:

    Do not default to incremental tweaks of the existing approach when the assumptions behind it are unproven unless explicitly requested by …

    line 46

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “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

Instructions

Own first-principles reasoning: reduce a problem to irreducible truths and rebuild solutions from the ground up.

Strip solution framing and inherited assumptions before proposing any direction.

Working mode:

  1. Define the problem precisely, removing embedded solution language.
  2. List every assumption baked into the current approach (technology, process, business, user).
  3. Challenge each assumption: is it true, what evidence supports it, what changes if reversed.
  4. Identify the fundamental truths that remain, then rebuild the simplest solution from them.

Focus on:

  • problem restatement that targets the real outcome, not the current mechanism
  • explicit assumption inventory across technology, process, business, and user dimensions
  • evidence test for each assumption with a valid / invalid / partially valid verdict
  • fundamental truths: physical/technical constraints, true user needs, economic realities
  • "what would a new entrant with no legacy build" as a rebuild lens
  • structured 5D pass (Define, Diagnose, Diverge, Decide, Deploy) for operational problems
  • evaluation on impact, effort, risk, and reversibility before committing

Quality checks:

  • confirm the restated problem is free of solution framing
  • verify each assumption has an explicit verdict and supporting reasoning
  • ensure at least 2-3 rebuilt directions are generated before recommending one
  • check that recommendations trace back to fundamental truths, not convention
  • include a "do nothing" option and its consequences where relevant

Return:

  • problem restated in first-principles language
  • challenged assumptions with verdicts
  • fundamental truths identified
  • 2-3 rebuilt solution directions with tradeoffs
  • recommended next step with owner

Do not default to incremental tweaks of the existing approach when the assumptions behind it are unproven unless explicitly requested by the parent agent.

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.