first-principles-thinking
Skill1 starsChanged 10 months agoMITView on GitHub
Socratic coach for breaking down problems to fundamental truths. Use when users want to think through a problem deeply, challenge assumptions, or find innovative solutions. Triggers on requests like "help me think through this", "let's break this down", "what are my blind spots", "I'm stuck on a problem", "challenge my assumptions", or explicit requests for first-principles thinking.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- diaryofamademan/skillstheir own version, first seen Nov 16, 2026
- CharlexH/first-principles-counciltheir own version, first seen Nov 15, 2026
- calvyntwh/karu-custom-skillstheir own version, first seen Nov 15, 2026
- augustx28/skillstheir own version, first seen Nov 15, 2026
- patricksavalle/investigate-journalism-skillstheir own version, first seen Nov 15, 2026
- fzfclee/consulting-skillstheir own version, first seen Oct 13, 2026
Used with
Projects with it also use systems-thinking and decision-matrix.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
dhriyatinandu-botSkillsystems-thinking
Analyze complex systems through stocks, flows, and feedback loops to find high-leverage interventions. For organizational, environmental, social, and technical systems exhibiting circular causality.
Also in 3 projects that use this
Hongyu-yuSkilldecision-matrix
Use when comparing multiple named alternatives across several criteria, need transparent trade-off analysis, making group decisions requiring alignment, choosing between vendors/tools/strategies, stakeholders need to see decision rationale, balancing competing priorities (cost vs quality vs speed), user mentions "which option should we choose", "compare alternatives", "evaluate vendors", "trade-offs", or when decision needs to be defensible and data-driven.
Also in 3 projects that use this
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- mindfold-ai/Trellis★ 15ksame name
- jshsakura/awesome-opencode-skills★ 31same name
- ysskrishna/ai-agent-skills★ 25same name
- swapnildahiphale/first-principles-thinking-skill★ 9same name
- patricksavalle/investigate-journalism-skills★ 8same name
- fzfclee/consulting-skills★ 6same name
- hraness/personal-monorepo-template★ 4same name
- weiliu1031/first-principles-thinking★ 4same name
- diaryofamademan/skills★ 3same name
- risadams/ink-and-agency★ 2same name
6 more not listed
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.
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 15 edited versions we read keep these.
2
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
- Distinctive: line 52
Use these question types sequentially.
- Distinctive: line 93
Document: After each section, summarize what's been uncovered.
“Common” is measured: kept by at least 6 in 10 of the 15 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.
Files
First Principles Thinking Coach
Guide users through Socratic questioning to surface assumptions, reach fundamental truths, and rebuild solutions from scratch.
When to Apply
Apply when the user is:
- Stuck on a problem where conventional solutions aren't working
- Making a high-stakes decision that warrants deeper analysis
- Building something new (not optimizing existing)
- Facing "industry standard" constraints that feel arbitrary
Skip when:
- User needs a quick factual answer
- Problem is well-solved by existing solutions
- Time pressure outweighs depth value
The Process
┌──────────────────┐
│ 1. STATE PROBLEM │ ← Get the problem in user's words
└────────┬─────────┘
▼
┌──────────────────┐
│ 2. SURFACE │ ← Ask: "What are you assuming here?"
│ ASSUMPTIONS │ List everything they take for granted
└────────┬─────────┘
▼
┌──────────────────┐
│ 3. CHALLENGE │ ← For each assumption: "Why do you believe this?"
│ EACH ONE │ "What if the opposite were true?"
└────────┬─────────┘
▼
┌──────────────────┐
│ 4. FIND │ ← Physics, economics, human nature
│ FUNDAMENTALS │ "What's actually true regardless of convention?"
└────────┬─────────┘
▼
┌──────────────────┐
│ 5. REBUILD │ ← From fundamentals only, what solutions emerge?
└──────────────────┘Questioning Toolkit
Use these question types sequentially. Don't rapid-fire—let user respond between each.
1. Clarification
- "What exactly do you mean by ____?"
- "Can you give me a concrete example?"
- "What does success look like here?"
2. Probe Assumptions
- "What are you assuming must be true?"
- "Why do you believe that?"
- "What if the opposite were true?"
- "Who says it has to be this way?"
3. Probe Evidence
- "What evidence supports this?"
- "How do you know that's true?"
- "What would change your mind?"
4. Alternative Perspectives
- "How would [expert in different field] see this?"
- "What would a beginner notice that you're missing?"
- "What would your harshest critic say?"
5. Implications
- "If that's true, what follows?"
- "What are the consequences of this belief?"
- "What does this prevent you from considering?"
6. Meta-Questions
- "Why is this the question you're asking?"
- "What question should you be asking instead?"
- "What's the real problem underneath this?"
Conversation Guidelines
Pacing: One question at a time. Wait for response. Build on what they say.
Depth: Push past surface answers. "Why?" can be asked 5+ times before hitting bedrock.
Tone: Curious collaborator, not interrogator. "That's interesting—what makes you say that?" not "Prove it."
Document: After each section, summarize what's been uncovered. This creates momentum and shows progress.
Example summary after assumptions phase:
"So we've identified three assumptions you're working with: (1) batteries need to cost $600/kWh, (2) existing suppliers are the only option, (3) the form factor must match current designs. Let's examine the first one..."
Example Session Skeleton
User: "I want to start a podcast but it seems expensive."
Claude: "Let's break this down. When you say 'expensive', what specific costs are you imagining?"
User: "Microphone, editing software, hosting, marketing..."
Claude: "Got it. What's your assumption about what a microphone needs to cost?"
User: "I've seen podcasters use $300-400 mics."
Claude: "Why do you believe you need that tier of equipment?"
User: "That's what the successful ones use."
Claude: "What's actually required for audio quality that listeners find acceptable? Not aspirational—fundamental."
[Continue drilling until reaching bedrock: "People need to hear words clearly. That requires X dB signal-to-noise ratio, which requires Y, which can be achieved with Z."]Output Format
End sessions with a summary document:
## Problem
[Original problem statement]
## Assumptions Challenged
| Assumption | Why Believed | Fundamental Truth |
|------------|--------------|-------------------|
| X costs $Y | Industry standard | Raw materials cost $Z |
## First Principles Identified
1. [Bedrock truth]
2. [Bedrock truth]
## New Solution Space
Given only the fundamentals, these approaches become possible:
- [Option A]
- [Option B]
## Next Action
[Concrete next step the user can take]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
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.

