diataxis/SKILL.md
Classify, validate, generate, and audit documentation using the Diátaxis framework. Enforces quadrant purity across tutorials, how-to guides, reference, and explanation. Use when creating documentation, reviewing docs, auditing doc sets, restructuring existing content, or when the user mentions Diátaxis, documentation structure, or doc quality.
npx skillsauth add keithpatton/diataxis-agent-skill diataxisInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill enforces the Diátaxis framework for documentation. Diátaxis identifies four distinct documentation types based on user needs: tutorials, how-to guides, reference, and explanation. Each serves a different purpose and must be written differently.
This skill operates in one of the following modes:
The mode must be inferred from the user request or explicitly stated. If the operating mode cannot be confidently inferred, ask for clarification before proceeding.
Refusal with explanation is a valid and successful outcome when quadrant purity cannot be maintained.
For generate mode, one of the following must be true before producing content:
Use this decision tree to classify content into exactly ONE quadrant:
| If content... | ...and serves user's... | ...then it belongs to... | |-------------------|---------------------------|--------------------------| | informs action | acquisition (study) | Tutorial | | informs action | application (work) | How-to Guide | | informs cognition | application (work) | Reference | | informs cognition | acquisition (study) | Explanation |
Two questions to ask:
A lesson that takes the learner through a practical experience. The instructor is responsible for the learner's success. Focus on doing, not explaining. Deliver results early and often. Minimize explanation. No choices or alternatives.
Directions that guide an already-competent user through a real-world problem. Assumes the reader knows what they want to achieve. Focused on work, not study.
Technical description of the machinery. Austere, neutral, accurate. Structure mirrors the thing being described. The user consults it, not reads it.
Discursive treatment that provides context, background, and answers "why". Makes connections, admits opinion, circles around the subject.
When classifying or flagging violations, always provide:
Example:
Classified as How-to (high confidence). Evidence: imperative verbs ("configure", "set up"), goal-oriented heading, absence of conceptual framing, assumes reader competence.
Flag these anti-patterns:
For detailed detection patterns and remediation, see references/anti-patterns.md.
This skill does not:
Diátaxis is the work of Daniele Procida. This skill encodes the Diátaxis framework for use by AI agents. For the authoritative source and complete documentation, see diataxis.fr.
Licensed under CC-BY-SA 4.0. Citation metadata available at the Diátaxis GitHub repository.
development
Maintainer-only workflow for handling GitHub Secret Scanning alerts on OpenClaw. Use when Codex needs to triage, redact, clean up, and resolve secret leakage found in issue comments, issue bodies, PR comments, or other GitHub content.
development
Maintainer workflow for OpenClaw releases, prereleases, changelog release notes, and publish validation. Use when Codex needs to prepare or verify stable or beta release steps, align version naming, assemble release notes, check release auth requirements, or validate publish-time commands and artifacts.
development
Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.
development
End-to-end Parallels smoke, upgrade, and rerun workflow for OpenClaw across macOS, Windows, and Linux guests. Use when Codex needs to run, rerun, debug, or interpret VM-based install, onboarding, gateway smoke tests, latest-release-to-main upgrade checks, fresh snapshot retests, or optional Discord roundtrip verification under Parallels.