What makes Circle of Competence notable?

998 downloads. Trust evidence available

How should Circle of Competence be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Circle of Competence

Activate when: user says 'should I enter this market / make this hire / invest in this domain,' 'am I qualified to evaluate this,' 'stay in your lane,' 'stic... Skill: Circle of Competence Owner: deciqai Summary: Activate when: user says 'should I enter this market / make this hire / invest in this domain,' 'am I qualified to evaluate this,' 'stay in your lane,' 'stic... Tags: latest:1.0.5 Version history: v1.0.5 | 2026-07-16T17:54:08.171Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/circle-of-competence.json) v1.0.4 | 2026-07-10T10:24:

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

998

Updated

Oct 11, 2026

Version

1.0.5

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 998 downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 998 downloadsadoption · observed Oct 11, 2026
Latest release: 1.0.5release · observed Jul 16, 2026
Handshake status: UNKNOWNsecurity

What changed

Newest first. Only changes that point at a source are listed.

Release 1.0.5release
Description tail link + agents machine-readable metadata line (deciqai.com/s/circle-of-competence.json)
Observed Jul 16, 2026 · source

Install and run

Setup complexity: low.

clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:circle-of-competence

Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-deciqai-circle-of-competence/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

143,640 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: circle-of-competence
description: "Activate when: user says 'should I enter this market / make this hire / invest in this domain,' 'am I qualified to evaluate this,' 'stay in your lane,' 'stick to what you know,' or is about to make a consequential decision in an unfamiliar domain. Do NOT activate when: the decision is small and explicitly framed as a learning bet; the domain is genuinely frontier for everyone and no one has competence yet. More: deciqai.com/c/circle-of-competence"
---

# Circle of Competence

## Overview

Every decision-maker has a domain where judgments are reliable (inside the circle) and one where they are not (outside). The discipline: identify the boundary, act decisively inside it, refuse to act outside it without explicit acknowledgment of elevated uncertainty.

Canonical source: Buffett (1996) — "The size of that circle is not very important; knowing its boundaries, however, is vital."

Composes with `dunning-kruger` (cognitive mechanism the circle counters), `metacognition` (self-monitoring the circle relies on), `first-principles` (only reliable inside the circle), `opportunity-cost` (refusal has a real cost — measure it), `probabilistic-thinking` (calibration training establishes the boundary).

## When to Use

- Evaluating an investment, acquisition, or partnership in an unfamiliar domain
- Deciding whether to enter a new market or product area
- A board member or investor is pushing you toward a domain outside your background
- A founder is asked to take on a function (legal, finance, sales) without the relevant track record
- Hiring senior roles in unfamiliar functions; allocating significant capital to a new business line
- Investing in, or building on, AI you don't truly understand — "should I fund/build this AI product," "is this AI hype or real," AI adoption pushed by FOMO

**Not when:** decision is small, reversible, and an explicit learning bet; refusing everything outside the current circle is genuinely worse than acting carefully; domain is genuinely frontier for everyone.

## Coaching Novices (Adaptive Front Door)

- **Engine mode:** user has a specific decision → run The Process directly.
- **Coach mode:** user is unfamiliar or has no concrete case → guide step by step.

In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.

1. One-line: before acting decisively, ask whether you have calibrated competence to evaluate the decision — if not, refuse or size the bet to the uncertainty.
2. Check fit: small and reversible → circle test is less load-bearing. Consequential and hard-to-reverse → essential.
3. Elicit the specific decision and the basis for evaluating it.
> **[WAIT — do not advance until user responds]**
4. One question at a time: have you been right in similar cases? Do you have calibration data? What's the boundary of your competence? If outside, what's the bet size?
> **[WAIT — do not advance until user responds]**
5. Close:

_meta.json

{
  "ownerId": "kn754b8sk22s8c6gjxt02bftbn88q7ye",
  "slug": "circle-of-competence",
  "version": "1.0.5",
  "publishedAt": 1784224448171
}

references/sources.md

# Sources — circle-of-competence

> *Primary sources for the [circle-of-competence](../SKILL.md) skill.*

- Buffett, W. E. (1996). "Letter to Shareholders." *Berkshire Hathaway Annual Report*. https://www.berkshirehathaway.com/letters/1996.html. The canonical articulation.
- Buffett, W. E. (1999). "Mr. Buffett on the Stock Market." *Fortune Magazine*, November 22, 1999. The contextual explanation during the tech bubble.
- Munger, C. T. (1994). "A Lesson on Elementary, Worldly Wisdom As It Relates To Investment Management and Business." USC Business School lecture. Reprinted in *Poor Charlie's Almanack* (2005). ISBN 978-1578645015.
- Marks, H. (2011). *The Most Important Thing: Uncommon Sense for the Thoughtful Investor.* Columbia Business School Publishing. ISBN 978-0231153683.
- Buffett, W. E. & Cunningham, L. A. (2013). *The Essays of Warren Buffett: Lessons for Corporate America* (4th ed.). The Cunningham Group. ISBN 978-1611634099.
- Greenwald, B. C. N. et al. (2001). *Value Investing: From Graham to Buffett and Beyond.* Wiley. ISBN 978-0471463399. The academic treatment.
- Pabrai, M. (2007). *The Dhandho Investor.* Wiley. ISBN 978-0470043899. A practitioner's application of the circle discipline.
- Deighton, J. (2002). "How Snapple Got Its Juice Back." *Harvard Business Review*, 80(1). https://hbr.org/2002/01/how-snapple-got-its-juice-back. The documented account of the Quaker-Snapple acquisition failure and Triarc revival.
- U.S. Federal Trade Commission (2023). "Keep your AI claims in check." Business guidance blog, Feb. 27, 2023. https://www.ftc.gov/business-guidance/blog/2023/02/keep-your-ai-claims-check. Regulatory caution against "AI washing" — over-claiming AI capability — relevant to circle discipline during the 2024-2026 AI boom.
- U.S. Securities and Exchange Commission (2024). "SEC Charges Two Investment Advisers with Making False and Misleading Statements About Their Use of Artificial Intelligence." Press release 2024-36, Mar. 18, 2024. https://www.sec.gov/newsroom/press-releases/2024-36. Enforcement documenting the gap between AI claims and competence in the 2024-2026 period.

examples/ai-hype-outside-the-circle-2024-2026.md

# Method in Action: Funding and Building AI Outside Your Circle (2024-2026)

> *Example for the [circle-of-competence](../SKILL.md) skill.*

The generative-AI boom that followed ChatGPT's late-2022 release created, by 2024-2026, the most widespread circle-of-competence stress test of the decade. Capital and attention flooded toward anything labeled "AI," and a large share of the people writing checks or building products could not honestly evaluate the technical claims they were betting on. The discipline here is not "avoid AI" — it is knowing the honest edge of your AI knowledge and either staying inside it or widening it deliberately, rather than letting hype substitute for competence.

This example walks a composite but representative decision — a generalist investor or non-technical founder weighing a bet on an "AI-powered" opportunity in 2025 — through the skill's own process.

**Step 1 — State the decision.** Domain(s): investing in (or building) a generative-AI product. Gut call: an enthusiastic yes — "AI is the biggest platform shift since mobile; I can't afford to miss it." Confidence: high. The high confidence paired with a novel domain is itself the first thing the skill flags.

**Step 2 — Circle test.** Framed honestly, the domain is not "software" or "technology in general" — it is *the specific technical and economic mechanics of large language models*: what a model can and cannot reliably do, the difference between a genuine moat and a thin wrapper over someone else's API, the real cost of inference and how gross margins behave at scale, and how quickly a foundation-model provider could ship the same feature and erase a startup's edge. Run against that honest domain, the test questions come back negative for a generalist:

- *Past decisions in this domain with measurable outcomes?* Reading about transformers or using ChatGPT daily is *exposure*, not a track record of correctly evaluating AI ventures through to outcome. Most 2024-2026 decision-makers had no such calibration data because the category was too young to have produced it.
- *Can you articulate the specific drivers experts would agree with?* The honest edge for a non-technical evaluator usually stops at the demo. Whether a product has defensibility — proprietary data, workflow lock-in, distribution — versus being a prompt wrapped in a UI that a model update or a well-funded incumbent can replicate is exactly the driver a generalist cannot reliably judge, and it is the driver that determines the outcome.
- *Would domain experts consider your evaluation credible?* Through this period, seasoned observers repeatedly warned that valuations and expectations had run ahead of demonstrated economics, and that many "AI" labels were marketing rather than substance. Regulators including the U.S. FTC publicly cautioned companies against overstating AI capabilities ("AI washing"), and the SEC brought enforcement actions in 2024 against firms for misleading AI-related claims. The expert

examples/buffetts-tech-stock-refusal-1990s.md

# Method in Action: Buffett's Tech-Stock Refusal, 1990s

> *Example for the [circle-of-competence](../SKILL.md) skill.*

The empirical demonstration of the circle-of-competence discipline is Warren Buffett's behavior throughout the 1990s technology boom. Buffett, who had been by then one of the most successful investors of the 20th century, refused to invest in technology companies on the basis that he could not evaluate them — and accepted the resulting opportunity cost as the price of the discipline.

The cost was real and visible. From 1990 to 1999, Microsoft's stock returned over 8,000%. Cisco returned over 75,000%. Dell returned over 30,000%. A Berkshire Hathaway portfolio with even modest allocations to these names would have produced returns substantially above Berkshire's actual performance over the same decade. Buffett's circle-of-competence discipline cost him — and Berkshire shareholders — billions of dollars in foregone gains.

Buffett was explicit about this in his 1999 *Fortune* essay (written near the peak of the tech bubble, when his "missed it" position was most visibly painful):

> "I have stayed out of certain industries because I lacked the ability to predict the long-term economics of any single business in those industries. I've also stayed out of industries that change too rapidly. ... The fact that I haven't found a tech company within my circle of competence doesn't mean others won't find tech companies within theirs. But I haven't.
>
> A great investment must be capable of being identified as great by the investor. The question 'Is Microsoft a great investment?' is, for me, the wrong question; the right question is 'Can I evaluate Microsoft well enough to know whether it's a great investment?' For me, the answer to the second question is no. For others, it may be yes. They have a circle of competence that includes technology; I do not.
>
> I will earn less than they earn during this period. I refuse to abandon the discipline that has produced my long-term results in order to capture short-term gains in an area where my judgment is not calibrated."
>
> — Buffett (1999), *Fortune*, November 22, 1999.

When the dot-com crash arrived in 2000-2002, the discipline vindicated itself in stark form. The Nasdaq dropped 78% from peak to trough. Many of the tech companies that had produced spectacular 1990s returns went to zero or near-zero. The investors who had abandoned circle-of-competence discipline during the boom paid the full cost during the bust. Buffett's foregone gains during the boom were preserved against the losses during the bust.

The longer-term lesson: the discipline of staying inside the circle costs visible upside in some periods and prevents catastrophic downside in others. The trade is not "circle discipline costs you" — it is "circle discipline trades upside variance for downside protection." For investors with long time horizons (Buffett's actual frame: forever), the protection-against-permanent-loss is struc

Github ReposUpdated 3d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW

Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular