Generate llms.txt and llms-full.txt files from site data sources.
npx skills add https://github.com/agentconfig/agentconfig.org --skill generate-llms
Please help me install this Agent Skill. Skill: generate-llms Source: https://github.com/agentconfig/agentconfig.org/tree/main/.github/skills/generate-llms Command: npx skills add https://github.com/agentconfig/agentconfig.org --skill generate-llms
Generates /llms.txt and /llms-full.txt files to help AI agents understand and navigate the agentconfig.org site content, keeping the index and full content up to date.
/llms.txt (TOC) and /llms-full.txt (complete content), plus per-page Markdown references like skills.md, agents.md, and mcp.md.Run the generation script to update site/public/llms.txt and site/public/llms-full.txt based on current data.
High-intent search queries and answers about installing and using this skill.
To generate llms.txt files automatically, this Skill reads from site data sources like pages.ts and writes outputs to site/public/llms.txt and llms-full.txt following the llmstxt.org standard. It assembles a coherent content bundle for AI agents to navigate your site.
The llms.txt file serves as a table of contents, while llms-full.txt contains the complete site content. This generation script creates both files plus per-page Markdown references like skills.md and agents.md to provide full context for AI navigation.
You should regenerate llms.txt before deployment or whenever site content changes, such as updates to pages, primitives, comparisons, or registries. This ensures AI agents receive accurate and current site context when reasoning about your content.
Yes, this Skill reads directly from TypeScript data sources like site/src/data/pages.ts, primitives.ts, and comparison.ts. It transforms these existing data structures into machine-readable Markdown files without requiring a separate content export process.
No external dependencies are required to generate llms-full.txt. The Skill operates using internal scripts to read site data sources and write the generated Markdown outputs directly to the site's public directory.
bendruckerProcess JSON data with jq filters for parsing, filtering, and transforming.
KSfakGenerate structured llms.txt files with absolute URLs from crawled site pages.
kquinslandScaffold Hugo tool bundles with front matter and integration metadata.
zzafergokConvert Markdown documents to HTML using marked.js, pandoc, gomarkdown, or static site generators.