What makes Antifragile notable?

1.2K downloads. Trust evidence available

How should Antifragile 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

Antifragile

Activate when: user asks whether their system/business/portfolio would survive a crisis; user says 'this has been fine for years but I'm nervous'; user wants... Skill: Antifragile Owner: deciqai Summary: Activate when: user asks whether their system/business/portfolio would survive a crisis; user says 'this has been fine for years but I'm nervous'; user wants... Tags: latest:1.0.5 Version history: v1.0.5 | 2026-07-16T17:51:45.314Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/antifragile.json) v1.0.4 | 2026-07-09T11:15:28.688Z | user Ref

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.2k

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. 1.2K 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: 1.2K downloadsadoption · observed Oct 11, 2026
Latest release: 1.0.5release · observed Jul 16, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:antifragile

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-antifragile/snapshot"

Documentation

CLAWHUB

132,522 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: antifragile
description: "Activate when: user asks whether their system/business/portfolio would survive a crisis; user says 'this has been fine for years but I'm nervous'; user wants to stress-test a plan against worst-case scenarios; user mentions Taleb, barbell strategy, via negativa, or skin in the game; user is deciding how to allocate across risky vs. safe options under high uncertainty.
Do NOT activate when: the decision is small and fully reversible with no meaningful downside; the system is simple, well-understood, and low-stakes. More: deciqai.com/c/antifragile"
---

# Antifragile

## Overview

Nassim Nicholas Taleb (2012) identified a third response to stress beyond fragile/robust: **antifragile** — systems that *gain* from disorder, with bounded downside and unbounded upside.

- **Fragile:** concave — absorbs small stress, breaks catastrophically at the tail. (Over-leveraged banks, just-in-time supply chains.)
- **Robust:** linear — unchanged by stress. (Physical infrastructure, traditional skills.)
- **Antifragile:** convex — improves under stress. (Evolution, the immune system, the restaurant industry as a whole.)

Core warning: **most modern complex systems are hidden-fragile** — stable only because the tail event hasn't arrived yet. Composes with `inversion`, `black-swan`, `expected-value-and-kelly`, `feedback-loops`.

## When to Use

- A system looks stable but may be hidden-fragile
- Designing a portfolio (financial, career, organizational) under uncertainty
- A "this can't happen" assumption is embedded in a strategic plan
- Recurrent small problems are suppressed rather than learned from
- A business depends on one AI/model vendor's API, pricing, or policy, or faces AI-native competition amid rapid AI capex and adoption shifts
- User says: "Taleb," "barbell strategy," "convex," "skin in the game," "via negativa"

**Not when:** decision is small and reversible; system is simple and low-stakes; you confuse high-variance with antifragile.

## Coaching Novices (Adaptive Front Door)

- **Engine mode:** user has a concrete system → run The Process directly.
- **Coach mode:** user is unfamiliar → 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: some things break under stress, some survive, **some get stronger** — most "stable" things are in the first category, just before stress arrives.
2. Check fit: small reversible decisions → not this lens.
3. Elicit their real system — what specifically are they stress-testing?
> **[WAIT — do not advance until user responds]**
4. Walk through The Process one step at a time with their input.
> **[WAIT — do not advance until user responds]**
5. Close by naming the specific design move (barbell / via negativa / optionality / skin-in-the-game) that fits their case.
> **[WAIT — do not advance until user responds]**

## The Process

**Step 1 — Classify exposure:** Under small / medium / ta

_meta.json

{
  "ownerId": "kn754b8sk22s8c6gjxt02bftbn88q7ye",
  "slug": "antifragile",
  "version": "1.0.5",
  "publishedAt": 1784224305314
}

references/sources.md

# Sources — antifragile

> *Primary sources for the [antifragile](../SKILL.md) skill.*

- Taleb, N. N. (2012). *Antifragile: Things That Gain from Disorder.* Random House. ISBN 978-1400067824. The founding text.
- Taleb, N. N. (2007). *The Black Swan: The Impact of the Highly Improbable.* Random House. ISBN 978-1400063512. The pre-2008 analysis of hidden fragility in the banking system.
- Taleb, N. N. (2018). *Skin in the Game: Hidden Asymmetries in Daily Life.* Random House. ISBN 978-0425284629. The follow-up on decision-maker / consequence alignment.
- Mandelbrot, B. B., & Hudson, R. L. (2004). *The (Mis)Behavior of Markets.* Basic Books. The mathematical foundation for fat-tail risk that Taleb builds on.
- Lo, A. W. (2017). *Adaptive Markets: Financial Evolution at the Speed of Thought.* Princeton University Press. Modern academic synthesis of adaptive/antifragile thinking in financial markets.
- United States. *Federal Aviation Act of 1958*, Pub. L. 85-726, 72 Stat. 731 (August 23, 1958). The law that consolidated U.S. air-safety authority into the FAA after the 1956 Grand Canyon collision — the historical anchor for aviation as an antifragile system.
- Andreessen Horowitz. "Who Owns the Generative AI Platform?" (a16z, January 2023). Widely-cited analysis of where durable value accrues in the generative-AI stack — arguing the undifferentiated application layer (thin wrappers) is fragile relative to model providers and infrastructure. Contemporary anchor for the 2024–2026 AI-business fragility example.
- Sequoia Capital. "AI's $600B Question" (Sequoia, 2024). On the gap between AI infrastructure/capex spend and application-layer revenue — the macro backdrop for classifying which AI businesses are hidden-fragile.

examples/1956-grand-canyon-collision-and-aviation-safety.md

# Method in Action: The 1956 Grand Canyon Collision and the Antifragile Aviation System (1956 → present)

> *Example for the [antifragile](../SKILL.md) skill.*

Commercial aviation is Taleb's cleanest example of a system that is antifragile *at the aggregate level* even though every individual unit in it is fragile. A single aircraft is concave to stress — one bad failure destroys it. But the *system* of air travel gains strength from each of those failures, because every crash is investigated and the findings are forced back into design and regulation. The 1956 Grand Canyon collision is the historical hinge where the United States built that feedback loop into law.

**Step 1 — Classify exposure.** On June 30, 1956, TWA Flight 2 and United Air Lines Flight 718 collided over the Grand Canyon in uncontrolled airspace, killing all 128 people aboard both planes. At the time, aircraft above the airways operated on a "see and be seen" basis with no unified traffic control. The individual aircraft were **fragile** (a tail event destroyed them completely). The question the disaster forced was about the *system's* exposure shape: would air travel merely absorb the loss (robust), or convert it into a permanently safer network (antifragile)?

**Step 2 — Identify hidden fragility.** The collision exposed a single point of failure hiding in plain sight: two federal bodies (the CAA and the CAB) with overlapping, under-funded authority, and vast stretches of high-altitude airspace with no positive control. The system had been "fine for years" only because traffic density had not yet forced the tail event. Rising post-war passenger volume was the stress that revealed it.

**Step 3 — Apply the design moves.** The response was structural, not cosmetic:
- **Skin in the game:** independent, mandatory, public accident investigation — the investigators bear the reputational cost of missing a cause, and findings cannot be buried by the operator.
- **Via negativa:** the Federal Aviation Act of 1958 dissolved the fragmenting overlap and *removed* the divided authority, consolidating airspace control under a single new agency (the FAA).
- **Optionality / redundancy:** mandatory positive control of high-altitude airspace, radar coverage, and later collision-avoidance systems (TCAS) — each a convex hedge with small routine cost and large tail payoff.

**Step 4 — Stress-test the claim.** The downside is *bounded* — one aircraft, one investigation. The upside *scales with disorder*: each accident feeds the corpus of known failure modes, so the marginal crash makes every subsequent flight safer. This is the signature of convexity, not high variance. Over the following decades the U.S. commercial fatal-accident rate fell by orders of magnitude even as flight volume rose — the integral effect of thousands of individual (fragile) failures processed by an (antifragile) system.

The lesson generalizes: **fragile parts can compose into an antifragile whole, but only if failures are

examples/ai-business-fragility-2024-2026.md

# Method in Action: Fragile vs. Antifragile AI Businesses (2024–2026)

> *Example for the [antifragile](../SKILL.md) skill.*

Between 2024 and 2026 a large cohort of "AI-native" companies was built on top of a handful of foundation-model APIs (OpenAI, Anthropic, Google, and open-weight families like Meta's Llama and Mistral). Many of these looked identical from the outside — a chat box, a workflow, a vertical assistant — but underneath they had opposite exposure shapes. The thin wrapper that resells one model's output is *fragile* to a single price, policy, or capability change it does not control. A company built with model optionality, proprietary data, and bounded-downside experiments is *antifragile*: the same volatility that kills the wrapper hands it customers, pricing power, and free capability upgrades. This example runs the anchor case through the skill's four steps.

**Step 1 — Classify exposure.** Take a "thin wrapper": a product whose core is a prompt plus one provider's API, with no proprietary data, no switching-cost moat, and margins set by that provider's token price.
- *Small stress* (a minor price cut by the provider): margins compress but the business survives — mildly concave.
- *Medium stress* (the provider ships a first-party feature that overlaps the wrapper's whole value proposition, or a rate-limit / policy change): the product's reason to exist collapses. This is not hypothetical — the running industry joke of 2023–2025 was that OpenAI could "kill your startup with a single release," and successive model and product launches repeatedly absorbed categories of thin wrappers.
- *Tail stress* (the provider bans the use case, deprecates the exact model the product is tuned around, or raises prices sharply): catastrophic loss.

The exposure curve is **concave** — small gains in the good case, uncapped loss in the bad case. That is the operational signature of a **fragile** system, and its apparent stability during 2024's funding boom was the *absence* of a stress test, not robustness to one.

Now take the antifragile counterpart: a company that routes across multiple models, owns proprietary data and workflow, and runs many small experiments. Under the *same* disorder, a price war between providers *lowers its costs*; a new frontier model *upgrades its product for free*; a competitor's collapse *sends it customers*. Its curve is **convex** — bounded downside, upside that scales with the very volatility that destroys the wrapper.

**Step 2 — Identify hidden fragility.** The fragile wrapper's fragilities are the classic checklist, all pointing at one counterparty:
- **Single point of failure / vendor concentration:** one model provider supplies the entire core function. Concentration on one vendor for a mission-critical function is the textbook SPOF.
- **No pricing power / commoditized input:** the wrapper cannot pass through cost shocks because a dozen near-identical competitors sit on the same API.
- **Untested "always been fi

Github ReposUpdated 1d 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.