agent entry point

Feed Koinara to your AI agent.

Koinara is built to be fetched without private workspace access. Start with the compact index, then choose a record page, raw Markdown file, RSS feed, or full archive depending on your context window.

Fast path

curl -L https://koinara.org/llms.txt

curl -L https://koinara.org/records/detached-head-before-push.md

curl -L https://koinara.org/llms-full.txt

wget -qO- https://koinara.org/rss.xml

Routes worth knowing

/llms.txt — compact agent index with every public record, summary, tags, and raw Markdown URL.
/llms-full.txt — one concatenated Markdown archive for agents that want a single context file.
/records/{slug}.md — raw Markdown for one record, including source URL and citation metadata.
/records/ — human-readable public record index.
/rss.xml — feed-aware route for recent records.
/sitemap.xml — crawler and discovery map.

How to cite a record

Prefer the stable HTML record URL as the citation target. Each record page now includes a Cite this record block with Markdown and plain-text forms. Raw Markdown endpoints include the same source URL and license metadata for agents that cannot fetch HTML reliably.

Content is CC BY-SA 4.0 unless a record says otherwise. If your style requires an access date, use the date your agent fetched the record.

If something looks wrong

If a record is stale, broken, or unsafe, open an issue at github.com/koinara/koinara-site. Koinara grows slowly on purpose; errata are welcome before new-record proposals.