# Agent instructions for robinkroonen.com

Personal dev blog of Robin Kroonen (Kroonen AI, Libre ecosystem). 1 published post(s). Everything is public and read-only: no authentication, registration, API keys, or rate-limit headers to negotiate.

## How to read this site

1. Discover: GET https://robinkroonen.com/llms.txt (curated index) or https://robinkroonen.com/posts.json (JSON index).
2. Search: GET https://robinkroonen.com/api/search?q=<query> — deterministic, returns JSON {results: [{slug, title, description, url, markdownUrl, score, snippet}]}.
3. Fetch content: GET /posts/<slug>/index.md for a single post, or https://robinkroonen.com/llms-full.txt for every post in one file.
4. Any HTML page can be requested with "Accept: text/markdown" to get markdown back (Content-Type: text/markdown, with an x-markdown-tokens estimate header).

## Conventions

- Cite the canonical HTML URL (https://robinkroonen.com/posts/<slug>/) when quoting, not the .md URL.
- Content is authored by Robin Kroonen unless a post states otherwise.
- Feeds: /rss.xml (RSS), /sitemap-index.xml (sitemap).
- Machine discovery: /.well-known/api-catalog (RFC 9727), /.well-known/ai-catalog.json (ARD), /.well-known/agent-skills/index.json (agent skills).