harness-docs

Generate a Docusaurus documentation site from repository sources and skill definitions.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/woink/harness-skills-plugin --skill harness-docs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-docs
Source: https://github.com/woink/harness-skills-plugin/tree/main/skills/harness-docs
Command: npx skills add https://github.com/woink/harness-skills-plugin --skill harness-docs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Generates a complete, structured documentation site from repository sources and skill metadata so teams don't have to write and keep reference pages, script docs, and changelogs manually. It performs cold reads of SKILL.md files, scripts, and README content and produces a coherent Docusaurus site plus a machine-readable generation report, enabling reproducible, file-based handoffs that survive context resets.

Core Features & Use Cases

  • Full-site scaffolding: On first run, scaffold a Docusaurus 3.x site with landing pages, per-skill reference pages, script docs, and blog changelog entries from git history.
  • Incremental regeneration: On re-runs, detect stale pages using last_generated frontmatter and only regenerate pages whose sources changed, minimizing churn.
  • Auditability and CI integration: Emit .harness/DOCS_REPORT.md with coverage, gaps, and git SHA, then commit changes to trigger documentation build workflows.
  • Robust error handling: Produce stub pages and detailed gap reports when SKILL.md frontmatter is malformed or sources are missing.

Quick Start

Generate a complete Docusaurus docs-site from the repository and write a .harness/DOCS_REPORT.md summarizing what was generated, skipped, and any gaps found.

Frequently Asked Questions about harness-docs

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

FAQPage Schema
How do I auto-generate Docusaurus documentation from a repository?

Auto-generate Docusaurus documentation by performing cold reads of SKILL.md files, scripts, and README content to scaffold a complete site with per-skill reference pages, script manuals, and changelog blog entries from git history.

Can I incrementally regenerate only stale documentation pages?

Incrementally regenerate stale documentation pages by checking last_generated frontmatter timestamps on re-runs, ensuring only pages whose underlying SKILL.md or script sources changed are updated to minimize churn.

Does auto-generating project documentation work with multi-skill repositories?

Auto-generating project documentation works specifically with Claude Code harness plugins and similar multi-skill repositories, reading optional .harness files alongside SKILL.md definitions to produce structured Docusaurus sites.

How do I track documentation coverage gaps and generation status?

Track documentation coverage gaps and generation status by emitting a machine-readable .harness/DOCS_REPORT.md file containing coverage metrics, missing sources, and the git SHA for CI integration and auditability.

What happens when SKILL.md frontmatter is malformed during documentation generation?

When SKILL.md frontmatter is malformed during documentation generation, the process produces stub pages and detailed gap reports rather than failing, ensuring robust error handling for missing or malformed sources.

Do I need git history to generate changelog blog entries for my docs site?

Git history is required to generate changelog blog entries for your docs site, as the Docusaurus scaffolding process reads commit history to produce chronological changelog posts alongside reference pages.