What makes Chesterton's Fence notable?

1K downloads. Trust evidence available

How should Chesterton's Fence 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

Chesterton's Fence

Activate when: someone says 'let's just remove this', 'why do we still have this rule?', 'this seems useless/outdated', 'nobody knows why this is here', new... Skill: Chesterton's Fence Owner: deciqai Summary: Activate when: someone says 'let's just remove this', 'why do we still have this rule?', 'this seems useless/outdated', 'nobody knows why this is here', new... Tags: latest:1.0.5 Version history: v1.0.5 | 2026-07-16T17:54:00.219Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/chestertons-fence.json) v1.0.4 | 2026-07-13T07:01:25.221

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.0k

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. 1K 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: 1K 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:chestertons-fence

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-chestertons-fence/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

144,575 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: chestertons-fence
description: "Activate when: someone says 'let's just remove this', 'why do we still have this rule?', 'this seems useless/outdated', 'nobody knows why this is here', new leadership restructuring without knowing the history, a developer deleting code whose purpose isn't documented, a regulator repealing a law without tracing its origin.
Do NOT activate when: the fence's history is fully documented and the documented purpose is confirmed obsolete (investigation already done); the reformer is the original builder and the rationale is fully understood. More: deciqai.com/c/chestertons-fence"
---

# Chesterton's Fence

## Overview

Before removing a rule, process, code path, or institution — you must understand why it was put there. Only when you can articulate the original purpose are you qualified to decide whether it still applies. Three components: (1) "I can't see the purpose" is evidence about you, not the fence; (2) investigation is mandatory, not optional; (3) demonstrated understanding is the prerequisite for change.

Composes with `survivorship-bias`, `second-order-thinking`, `feedback-loops`, `first-principles`.

## When to Use

- A rule, process, code path, or practice is proposed for removal
- New leadership restructuring an organization with unfamiliar practices
- A developer "cleaning up" code whose purpose isn't documented
- A regulator or legislator repealing existing protections
- Someone says "why is this here?", "let's just remove this", "this seems useless"
- An AI-assisted rewrite/refactor proposes deleting an "ugly" guardrail, edge-case branch, validation, or manual review gate the model calls redundant

**Not when:** fence history is fully documented and purpose is confirmed obsolete; reformer is the original builder with full rationale understood.

## Coaching Novices (Adaptive Front Door)

- **Engine mode:** user has a concrete fence-removal case → 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 removing a rule/code/process whose purpose you can't articulate, investigate — your inability to see the purpose is data about you, not the rule.
2. Check fit: if the fence's history is fully documented and the purpose is clearly obsolete, the investigation is already done.
3. Elicit the specific fence and proposed removal. What's being removed? Who proposes it? Why?
> **[WAIT — do not advance until user responds]**
4. One question at a time: when was the fence put there? by whom? what problem was it solving? does that problem still exist? are there other defenses?
> **[WAIT — do not advance until user responds]**
5. Close: investigation summary (purpose found / not found) + decision (remove / keep / modify) + documentation so the next reformer can read the history.
> **[WAIT — do not advance until user respo

_meta.json

{
  "ownerId": "kn754b8sk22s8c6gjxt02bftbn88q7ye",
  "slug": "chestertons-fence",
  "version": "1.0.5",
  "publishedAt": 1784224440219
}

references/sources.md

# Sources — chestertons-fence

> *Primary sources for the [chestertons-fence](../SKILL.md) skill.*

- Chesterton, G. K. (1929). *The Thing.* Sheed & Ward. ISBN 978-1602068191 (reprint). The original formulation, chapter "The Drift from Domesticity."
- Hayek, F. A. (1973). *Law, Legislation and Liberty, Volume 1: Rules and Order.* University of Chicago Press. The spontaneous-order theoretical extension.
- Spolsky, J. (2000). "Things You Should Never Do, Part I." *Joel on Software* (joelonsoftware.com). The software-engineering application.
- Stiglitz, J. E. (2010). *Freefall: America, Free Markets, and the Sinking of the World Economy.* W. W. Norton. ISBN 978-0393075960. The Glass-Steagall application.
- Bourdieu, P. (1972). *Outline of a Theory of Practice.* Cambridge University Press. ISBN 978-0521291644. The anthropological foundation.
- Kotter, J. P. (1996). *Leading Change.* Harvard Business School Press. ISBN 978-0875847474. Organizational-change methodology.
- OECD (2008). *Building an Institutional Framework for Regulatory Impact Analysis (RIA): Guidance for Policy Makers.* The regulatory institutionalization.
- Shapiro, J. (2001). *Mao's War Against Nature: Politics and the Environment in Revolutionary China.* Cambridge University Press. ISBN 978-0521786805. The Great Sparrow Campaign / ecological-fence application.
- GitHub. *GitHub Copilot* product documentation (github.com/features/copilot). The AI code-generation/refactor tooling whose 2024–2026 adoption drives the AI-rewrite fence-removal pattern.
- Cursor (Anysphere). Product documentation (cursor.com / docs.cursor.com). AI-assisted rewrite/refactor tooling for the 2024–2026 AI-rewrite-wave application.

examples/1958-great-sparrow-campaign-and-the-ecological-fence.md

# Method in Action: The Great Sparrow Campaign and the Ecological Fence (1958–1962)

> *Example for the [chestertons-fence](../SKILL.md) skill.*

The Four Pests Campaign in China is the clearest ecological case of a fence removed without investigating its purpose. The "fence" was not a rule or a code path — it was a species. The Eurasian tree sparrow was an unnoticed, load-bearing component of the agricultural ecosystem, and its elimination is a textbook failure to run The Process before removal.

**Step 1 — Identify.** In 1958, as part of the Great Leap Forward, the sparrow was named one of the "Four Pests" (alongside rats, flies, and mosquitoes) and marked for extermination. The stated reason: sparrows eat grain seed, so killing them would raise harvests. The proposer was the central campaign itself; the time pressure was ideological and total — nationwide mobilization, no deliberation. Citizens across China banged pots and drums to keep sparrows airborne until the birds dropped dead from exhaustion; nests were destroyed and eggs broken.

**Step 2 — Investigate origin (the step that was skipped).** The relevant "fence" question was never asked: *what does the sparrow do in this system besides eat grain?* The answer, well established in ornithology, is that the sparrow's diet is largely insects, especially during the breeding season when it feeds its young almost entirely on insect larvae — including locusts and the pests that attack rice. The bird was a natural check on the insect populations that devastate crops. The campaign treated the sparrow "as a senseless monstrosity that has somehow sprung up in his path," in Chesterton's phrase, rather than investigating the ecological role it had long played.

**Step 3 — Judge current applicability (what investigation would have revealed).** The original "problem" — sparrows eating some grain — was real but small relative to the defense the sparrow provided against insect infestation. Removing the sparrow did not remove the load it bore; it transferred that load to nothing. There was no redundant fence: no other predator was ready to suppress the locusts at the same scale. The cost of keeping the sparrow was a modest loss of seed. The cost of removing it was the failure mode the fence had been silently preventing.

**Step 4 — Decide (the actual decision, and its reversal).** The campaign decided to remove the fence. Within two years the consequence appeared: with their natural predator gone, locust and insect populations exploded and stripped crops across the country. The ornithologist Tso-hsin Cheng and other scientists warned the leadership, and in 1960 the sparrow was quietly struck from the list of pests and replaced with the bed bug. The removal had been reversed — but reinstalling the fence was not free. Sparrow populations could not be restored on command; China reportedly imported sparrows to rebuild the population. The ecological damage compounded alongside drought and policy failure into the

examples/ai-rewrite-wave-deleting-guardrails-2024-2026.md

# Method in Action: The AI-Rewrite Wave and the Guardrails That Encoded Hard-Won Knowledge (2024–2026)

> *Example for the [chestertons-fence](../SKILL.md) skill.*

Between 2024 and 2026, coding assistants (GitHub Copilot, Cursor, Claude Code, and similar tools) made it cheap to regenerate large swaths of a codebase on demand. A recurring failure pattern emerged: a team, or an agent acting on a "clean this up / rewrite it with AI" prompt, deletes an "ugly" null check, a seemingly redundant validation rule, a manual review gate, or a defensive edge-case branch — because neither the human nor the model can see why it exists. The line has no comment, the original author is gone, and it looks like clutter. That is the Chesterton's Fence situation in its purest modern form: the fence is a line of code or a human-in-the-loop step that quietly encodes edge-case, security, or compliance knowledge that no longer lives in anyone's head. This walkthrough runs the generic pattern through The Process.

**Step 1 — Identify.** Fence: an "odd-looking" guardrail — e.g. a special-case branch that rejects a specific malformed input, a rate limit that seems too conservative, a manual approval step before a payout, or a validation that duplicates something the framework "already does." Proposed change: delete it during an AI-assisted refactor / rewrite. Proposer: a developer prompting a coding agent to "simplify," "modernize," or "rewrite this module," or an autonomous agent optimizing for fewer lines and passing tests. Stated reason: "the model flagged this as dead/redundant code," "it's not covered by any test," "it reads like legacy cruft." Time pressure: high — the whole appeal of the AI-rewrite is speed, and reviewing a large model-generated diff line-by-line is exactly the slow step teams are trying to skip.

**Step 2 — Investigate origin (the step the speed pressure skips).** The fence question: *why was this branch or gate added?* Investigation methods map directly onto Step 2's toolkit. `git blame` the line to the original commit and read the linked PR, issue, or incident — guardrails like this are very often the scar tissue of a past outage, a security disclosure, or a regulator's finding. Check whether the branch corresponds to a compliance requirement (e.g. a payment gate mandated by policy, a data-handling rule tied to privacy law) that is enforced by *convention in code* rather than by an obvious external checklist. The core trap of the 2024–2026 wave: an LLM sees the code's *syntax* and its test coverage, but it cannot see the *incident history* — the reason the fence exists lives in git history, ticket systems, and institutional memory that are not in the model's context window. "No test covers it" and "the model can't explain it" are, per the skill, evidence about the observer, not proof the fence is useless.

**Step 3 — Judge current applicability.** Does the original problem still exist? Often yes: the malformed input still arrives, the attack stil

Github ReposUpdated 2d 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.