Public API

A directory bots can read

Search the catalog, filter it, or keep a local copy continuously in sync. The API returns JSON, requires no key, supports cross-origin requests, and is cached at Cloudflare's edge for 60 seconds.

GET https://api.botdirectory.ai/api/bots

Browse and search

Results default to newest first. Use any combination of these query parameters:

Parameter	Meaning	Default
`q`	Search names, prompts, contributors, categories, and integrations	—
`category`	Exact category, case-insensitive	—
`integration`	Exact integration, case-insensitive	—
`page`	Page number	`1`
`limit`	Results per request; maximum 100	`25`
`sort`	`newest` or `name`	`newest`

GET https://api.botdirectory.ai/api/bots?q=slack&category=Ops&limit=25&sort=newest

Continuously sync new bots

Cursor mode is append-safe. It returns bots oldest-to-newest, so a client can save each page and append later additions without page numbers shifting underneath it.

Start with cursor=start.
Store sync.nextCursor from the response.
Call the returned links.next URL later and append any bots returned.

GET https://api.botdirectory.ai/api/bots?cursor=start&limit=100

{
  "bots": [ ... ],
  "sync": {
    "returned": 100,
    "hasMore": true,
    "nextCursor": "WyIyMDI2LTA4..."
  },
  "links": {
    "next": "https://api.botdirectory.ai/api/bots?cursor=WyIyMDI2LTA4...&limit=100"
  }
}

Keep the same search and filter parameters when reusing a cursor.

Send this page to a bot

This page contains the complete usage contract and can be read without JavaScript. Give a bot https://botdirectory.ai/api/ and ask it to browse the directory or maintain a cursor-based local mirror.

Raw feed

For a small one-request mirror, fetch https://botdirectory.ai/api/bots.json. It returns the complete catalog without pagination. Prefer cursor mode for ongoing synchronization.