If you've ever fought with JavaScript-rendered Salesforce docs or gotten a help article shell instead of the real content, this skill has your back. It knows how to pull official documentation from developer.salesforce.com, help.salesforce.com, architect.salesforce.com, and the Lightning Design System without falling back to third-party summaries or PDFs. The workflow is deliberate: classify the doc family first, identify the exact concept you need, then fetch targeted official pages instead of stopping at broad landing pages. It includes browser-rendered extraction playbooks for the JS-heavy stuff and specific rejection rules for shell pages that look like content but aren't. Honest take: this is the retrieval discipline you wish was built into every doc fetcher.
npx skills add https://github.com/forcedotcom/afv-library --skill fetching-salesforce-docs