reversa-docs-publisher

Generates and publishes an offline-capable HTML documentation mini-site with navigation, seals, and telemetry.

Updated Jun 12, 2026
One-click install
npx skills add https://github.com/lacsousa/mediclaw-system --skill reversa-docs-publisher-lacsousa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: reversa-docs-publisher
Source: https://github.com/lacsousa/mediclaw-system/tree/main/.agents/skills/reversa-docs-publisher
Command: npx skills add https://github.com/lacsousa/mediclaw-system --skill reversa-docs-publisher-lacsousa

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? It consolidates HTML pages produced by multiple documentation agents into a single coherent mini-site that works offline and via file://, eliminating broken links, missing vendor libraries, and CORS failures from local fetch calls. ## Core Features & Use Cases - Site assembly and injection: Generates index.html with a hero seal, injects a mini-seal and navigation menu retroactively into all existing pages, and embeds intermediate JSON data into a single window.RV_DATA bundle. - Offline vendor bundling: Downloads pinned versions of Three.js, D3, and Highcharts (with CDN fallbacks) into assets/vendor/ so no page depends on external CDNs. - Auto-discovery and validation: Scans auxiliary HTML outputs from other agents via meta tags, validates relative links, generates placeholder pages for omitted sections, and runs an HTTP smoke test before reporting final telemetry. - Use Case: After running the Reversa Docs pipeline agents (mapper, analyst, storyteller), invoke the publisher to produce a navigable index.html that a stakeholder can open with a double click, fully offline. ## Quick Start Run the reversa-docs-publisher agent after the other Reversa Docs agents have generated their pages, and ask it to assemble and publish the final mini-site in _reversa_docs/.

Frequently Asked Questions about reversa-docs-publisher

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

FAQPage Schema
How do I make a generated HTML documentation site work offline?▼

Download all third-party libraries such as Three.js, D3, and Highcharts into a local assets/vendor directory and reference them with relative script tags. Embed JSON data inline via a data.js file instead of using fetch, which browsers block on file:// origins.

Why does fetch fail when opening HTML files with a double click?▼

Browsers block fetch requests to local files because file:// pages have a null origin, triggering CORS restrictions. The fix is to embed data directly in a JavaScript file, such as window.RV_DATA, loaded via a script tag before the page logic runs.

How do I integrate HTML pages generated by multiple AI agents into one site?▼

Use a publisher step that discovers pages via meta tags like reversa-category, injects a shared navigation block and branding into each page, generates an index.html linking everything, and validates all relative links before finalizing.

Which Three.js version still supports OrbitControls as an IIFE script?▼

Three.js 0.147.0 is the last release shipping OrbitControls in IIFE format under examples/js/controls. From 0.150 onward only ES module versions exist, requiring an importmap and module scripts.

What happens if CDN downloads fail during site generation?▼

The publisher tries each library's primary URL, then ordered fallbacks across unpkg, jsdelivr, and cdnjs. If all fail, it records the missing vendor in state telemetry, renders a placeholder warning on affected pages, and continues publishing.