---
title: "ai-catalog.json"
type: "glossary-term"
category: "Discovery for agents"
canonical_url: "https://www.graphcanon.com/glossary/discovery-for-agents/ai-catalog"
updated: "2026-07-09"
---

# ai-catalog.json

_Also known as: Agent-readable dataset, ARD_

**ai-catalog.json is a well-known JSON manifest that advertises a site's agent-facing surfaces - its API, dataset, and MCP endpoint - in one discoverable place.**

**ai-catalog.json** is a manifest served under `/.well-known/` that declares, in one machine-readable document, how an AI system should consume a site: where the API lives, how to page it, what dataset is available, and how to reach any MCP server.

It is a discovery entry point - the single file an agent can fetch first to learn everything it needs to work with a site as data.

## In GraphCanon

GraphCanon serves "/.well-known/ai-catalog.json" describing its JSON API, corpus dataset, and MCP endpoint, so an agent can bootstrap from one URL instead of scraping.

## See also

- [ai-catalog.json](/.well-known/ai-catalog.json)
- [JSON API](/api)
- [MCP server docs](/docs/mcp)

## Related terms

- [Structured data (JSON-LD)](/glossary/discovery-for-agents/structured-data.md)
- [Model Context Protocol (MCP)](/glossary/ai-and-llms/model-context-protocol.md)

[Discovery for agents](/glossary/discovery-for-agents.md) · [All glossary terms](/glossary.md)

---

**Machine-readable endpoints**

- JSON: [`/api/graphcanon`](/api/graphcanon)
- LLM index: [/llms.txt](/llms.txt)
- Full corpus: [/llms-full.txt](/llms-full.txt)

_GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_