thinking-circle-of-competence
Skill2 starsChanged 10 months agoMITView on GitHub
Know the boundaries of your expertise and operate within them. Use when evaluating opportunities, making decisions outside your domain, or assessing when to defer to experts.
In this repository38 other files it gives its agents.
- thinking-archetypesSkill
- thinking-bayesianSkill
- thinking-bounded-rationalitySkill
- thinking-cynefinSkill
- thinking-debiasingSkill
- thinking-dual-processSkill
- thinking-effectuationSkill
- thinking-feedback-loopsSkill
- thinking-fermi-estimationSkill
- thinking-first-principlesSkill
- thinking-five-whys-plusSkill
- thinking-inversionSkill
- thinking-jobs-to-be-doneSkill
- thinking-kepner-tregoeSkill
- thinking-leverage-pointsSkill
- thinking-lindy-effectSkill
- thinking-map-territorySkill
- thinking-margin-of-safetySkill
- thinking-model-combinationSkill
- thinking-model-routerSkill
- thinking-model-selectionSkill
- thinking-occams-razorSkill
- thinking-oodaSkill
- thinking-opportunity-costSkill
- thinking-pre-mortemSkill
- thinking-probabilisticSkill
- thinking-red-teamSkill
- thinking-regret-minimizationSkill
- thinking-reversibilitySkill
- thinking-scientific-methodSkill
- thinking-second-orderSkill
- thinking-socraticSkill
- thinking-steel-manningSkill
- thinking-systemsSkill
- thinking-theory-of-constraintsSkill
- thinking-thought-experimentSkill
- thinking-trizSkill
- thinking-via-negativaSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
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
- Distinctive: line 78
Should delegate or deeply learn before deciding
- Distinctive: line 118
Look at past decisions:
- Distinctive: line 240
Start at the edge, not outside
- Distinctive: line 251
Avoid feedback on decisions
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.
Files
Circle of Competence
Overview
The Circle of Competence, articulated by Warren Buffett and Charlie Munger, emphasizes knowing the boundaries of your genuine expertise. The key insight isn't about having a large circle—it's about knowing precisely where your circle ends. Operating within your circle leads to better decisions; operating outside it without recognizing it leads to costly mistakes.
Core Principle: "Know what you don't know. The boundaries of your circle are more important than its size." — Warren Buffett
When to Use
- Evaluating new opportunities or projects
- Making decisions in unfamiliar domains
- Assessing whether to delegate or learn
- Investment or resource allocation decisions
- Taking on new responsibilities
- Advising others on topics
- Hiring or team composition decisions
Decision flow:
Decision to make? → yes → Inside your circle? → yes → Proceed with confidence
↘ no → Delegate, learn, or pass
↘ no → Not applicableThe Three Zones
Zone 1: Inside Your Circle
True competence through deep experience
Characteristics:
- You've made decisions here repeatedly
- You've seen failure modes firsthand
- You can predict second-order effects
- You know what you don't know within this area
- You can teach others the nuances
Example: Senior backend engineer
Inside circle:
- API design patterns that scale
- Database optimization strategies
- When to use caching vs. not
- Common failure modes in distributed systems
- Debugging production issuesZone 2: Edge of Your Circle
Familiar but not expert
Characteristics:
- You know the basics
- You can have informed conversations
- You might miss edge cases
- You need to verify your assumptions
- You should seek review from experts
Example: Same backend engineer
Edge of circle:
- Frontend performance optimization
- Basic security practices
- Cloud cost optimization
- Team management fundamentalsZone 3: Outside Your Circle
Dangerous territory
Characteristics:
- Knowledge is superficial or outdated
- You don't know what you don't know
- Decisions feel confident but are often wrong
- High risk of Dunning-Kruger effect
- Should delegate or deeply learn before deciding
Example: Same backend engineer
Outside circle:
- Mobile app development
- Machine learning model tuning
- Legal/compliance decisions
- Financial forecastingMapping Your Circle
Step 1: List Your Domains
What areas do you have experience in?
- Professional skills
- Industry knowledge
- Technical domains
- Business functions
Step 2: Assess Depth Honestly
For each domain, ask:
| Question | Inside | Edge | Outside |
| Could I teach this to an expert? | ✓ | ||
| Have I made real decisions here? | ✓ | ✓ | |
| Do I know the failure modes? | ✓ | ||
| Can I predict second-order effects? | ✓ | ||
| Do I know what I don't know here? | ✓ | ✓ | |
| Is my knowledge current? | ✓ |
Step 3: Be Brutally Honest
Common self-deceptions:
- "I read a lot about it" ≠ competence
- "I understand the concepts" ≠ can execute
- "I did it once years ago" ≠ current competence
- "I'm smart, I can figure it out" ≠ expertise
Step 4: Verify with Track Record
Look at past decisions:
- In areas you consider "inside," were you right?
- In areas you consider "edge," did you make mistakes?
- This calibrates your self-assessment
Operating Within Your Circle
When Inside: Act with Confidence
✓ Make decisions directly
✓ Move quickly
✓ Trust your intuition (it's trained)
✓ Teach and mentor others
✓ Push back on outside opinions if warrantedWhen at Edge: Proceed with Caution
→ Seek input from those with deeper expertise
→ Validate assumptions before acting
→ Build in more margin for error
→ Document reasoning for review
→ Use this as learning opportunityWhen Outside: Delegate or Learn
Option A: Delegate
- Find someone with this in their circle
- Trust their judgment
- Don't override without strong reason
Option B: Learn First
- Invest significant time (months/years)
- Get hands-on experience
- Make small decisions first, learn from mistakes
- Gradually expand circle
Option C: Pass
- Some opportunities aren't for you
- "I don't know enough" is valid
- Opportunity cost of learning may be too highCommon Traps
Trap 1: Circle Creep
Your circle in one area doesn't extend to adjacent areas:
Inside: iOS development
Doesn't mean inside: Android development
Doesn't mean inside: iOS design
Doesn't mean inside: iOS project managementTrap 2: Stale Expertise
Circles shrink if not maintained:
2015: Expert in jQuery
2024: jQuery knowledge is inside, but modern frontend is edge/outsideTrap 3: Confidence Misread as Competence
Feeling confident ≠ being competent:
Dunning-Kruger peak: Know just enough to feel expert
Actually expert: Know enough to know how much you don't knowTrap 4: Smart Person Syndrome
General intelligence doesn't expand circles:
Being smart at X doesn't make you competent at Y
Many smart people fail at investments, businesses, etc.
because they operate outside their circle confidentlyApplication Examples
Technical Decisions
Question: Should we adopt GraphQL?
Self-assessment:
- Have I built and maintained GraphQL at scale? No → Outside
- Do I know the failure modes? No → Outside
- Have I seen it succeed/fail in similar contexts? Partially → Edge
Decision: Consult with team members who have deep GraphQL experience,
or run small pilot before committingCareer Decisions
Question: Should I become a manager?
Self-assessment:
- Have I led teams before? A little → Edge
- Do I know management failure modes? Not really → Outside
- Can I predict what makes a good manager? Vaguely → Edge
Decision: Don't assume IC success transfers; seek mentorship,
start with small team, treat as learning opportunityBusiness Decisions
Question: Should I invest in this startup?
Self-assessment:
- Do I understand this market? Superficially → Edge
- Can I evaluate the technology? No → Outside
- Do I know startup failure modes? Generally → Edge
- Have I made successful startup investments? No → Outside
Decision: This is outside my circle; either pass or find
co-investors who have this in their circleExpanding Your Circle
The Right Way
- Start at the edge, not outside
- Make small, reversible decisions
- Get feedback on those decisions
- Learn from mistakes in low-stakes situations
- Build pattern recognition over time
- Graduate to larger decisions as track record develops
The Wrong Way
- Jump straight to big decisions in new domain
- Assume competence transfers
- Learn only theory without practice
- Avoid feedback on decisions
- Never acknowledge mistakes
Verification Checklist
- Identified which zone this decision falls in
- If edge/outside: acknowledged uncertainty explicitly
- If outside: identified who has this in their circle
- If proceeding outside circle: limited downside exposure
- Honest about track record in this domain
- Not conflating confidence with competence
Key Questions
- "Would I bet significant money on my judgment here?"
- "Have I made similar decisions successfully before?"
- "What don't I know that an expert would know?"
- "If I'm wrong, how would I know?"
- "Who has this in their circle that I could consult?"
- "Is my knowledge here current or stale?"
Buffett's Reminder
"What counts for most people in investing is not how much they know, but rather how realistically they define what they don't know."
The advantage comes not from having the biggest circle, but from staying inside whatever circle you have—and knowing exactly where the boundary is.
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.

