crawl-docs

Overview

Crawls a documentation site, streams its content through an agent summariser, and stores the result under .ouro/technologies/. The command also updates the curated docs map so downstream tooling can link to the new reference.

Usage

ouro crawl-docs <package-name> <website-url> [--agent-model <model>] [--max-pages <n>]

Key Options

• --agent-model <model> – Override the model used while synthesising the crawled documentation.
• --max-pages <n> – Limit the number of pages fetched from the target site.

Inputs & Outputs

• Inputs: package identifier, documentation base URL, optional model overrides.
• Outputs: Markdown stored at .ouro/technologies/<package>.md plus an updated .ouro/technologies/MAP.md entry.

When to Use

• Capture authoritative vendor documentation for a library that lacks a comprehensive README.
• Refresh existing technology notes after documentation revamps.

Follow-up Checklist

• Inspect the generated Markdown for navigation artefacts or broken links.
• Commit both the technology doc and the updated docs map.