Developers & AI
The same structured, server-rendered pages that humans read are designed to be legible to machines. Here is how to access the corpus programmatically.
MCP server
Model Context Protocol tools and resources — lookup by reference, semantic search, fetch passage, and list structure — so any MCP-capable agent can use the canon.
Claude Skill
A packaged skill that teaches an agent how to navigate, cite, and quote the Tipiṭaka correctly.
Public read API
A documented REST/JSON endpoint for fetching texts, translations, and search results.
Open data export
Bulk download of the texts (with licensing) so researchers and models can ingest the full canon.
Quick reference
Placeholder endpoints — the API is implemented later.
# Fetch a passage as JSON
GET /api/v1/text/sutta/dn/silakkhandha-vagga/dn1
# Semantic search
GET /api/v1/search?q=overcoming+anger&mode=semantic
# Machine-readable index for AI crawlers
GET /llms.txt