openclaw-refactor-docs

Refactors OpenClaw documentation pages with source-audited fact preservation and verification.

1|Updated Jun 9, 2026
One-click install
npx skills add https://github.com/JEROME-PRAKASH-L/openclaw --skill openclaw-refactor-docs-jerome-prakash-l
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openclaw-refactor-docs
Source: https://github.com/JEROME-PRAKASH-L/openclaw/tree/main/.agents/skills/openclaw-refactor-docs
Command: npx skills add https://github.com/JEROME-PRAKASH-L/openclaw --skill openclaw-refactor-docs-jerome-prakash-l

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Rewriting a large documentation page risks silently dropping accurate behavior details like config defaults, precedence rules, and error handling. This Skill provides a structured workflow to refactor OpenClaw docs pages while auditing every fact against source code so nothing correct is lost. ## Core Features & Use Cases - Fact Inventory and Audit: Builds a working inventory of config fields, commands, defaults, limits, and troubleshooting details from the old page, then decides to keep, move, or delete each one with source-backed justification. - Page-Type Classification and Splitting: Classifies the target page as topic, guide, reference, or troubleshooting, and plans where exhaustive contracts and symptom-driven material should move. - Source-of-Truth Verification: Checks behavior-sensitive claims against schemas, CLI implementations, runtime source, tests, and generated docs rather than inferring from names. - Use Case: When asked to shorten an overloaded page like docs/plugins/codex-harness.md, use this Skill to split reference tables into a dedicated page, preserve every verified behavior fact, and run docs validation checks. ## Quick Start Refactor the OpenClaw docs page at docs/plugins/codex-harness.md into a concise guide and move exhaustive config details to a reference page.

Frequently Asked Questions about openclaw-refactor-docs

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I refactor a documentation page without losing information?

Build an inventory of every behavior fact in the old page first, then decide for each fact whether to keep it, move it to a named destination page, or delete it with source-backed proof it is obsolete. Compare old and new versions before finishing.

How to split a long docs page into guide and reference pages?

Classify the page type first, then keep the 80/20 path, decisions, and common failures in the main guide while moving exhaustive fields, defaults, enums, and rare debugging details to a reference page. Link from the point where readers need the deeper detail.

Does this skill work with the openclaw-docs skill?

Yes, it builds directly on openclaw-docs and instructs you to read that skill first for page types, style, examples, navigation, and verification guidance. This skill adds the rewrite workflow on top of that standard.

How are behavior claims in docs verified against source code?

Each behavior-sensitive claim is checked against the nearest authoritative source: schemas and manifests for config fields, CLI implementations and tests for commands, and runtime source for lifecycle, permission, and timeout behavior. Existing docs count only as secondary evidence.

What validation checks run after a docs refactor?

Run the smallest reliable checks for the touched surface: pnpm docs:list, git diff --check, targeted oxfmt checks, docs MDX and link checks, i18n glossary checks when labels changed, and generated-doc checks when schemas or generated baselines are touched.