Use AgentGid from your AI
JSON, RSS and llms.txt
AgentGid's data as plain files anyone can fetch without signing in. JSON for agents, APIs, deals and MCP servers, RSS feeds, llms.txt and Markdown copies of the Docs.
Everything below is free and needs no sign-in. For an AI agent, the MCP server is usually easier; these files are for scripts, feed readers and assistants without MCP.
JSON #
| File | What is in it |
|---|---|
| /api/agents.json | Every ranked agent: rank, Gid Score, downloads, stars, latest release, and the date of each number (data_as_of, data_out_of_date) |
| /api/apis.json | APIs for agents with official prices and free tiers |
| /api/deals.json | Official promotions, credits, student and startup programs |
| /api/resets.json | Claude Code and Codex usage-limit resets |
| /api/real-world.json | What public GitHub repos actually configure |
| /api/mcp.json | MCP servers |
Each agent page also links its daily numbers as a CSV file. If you use the data, a link back is appreciated.
RSS #
- News: releases, outages, price changes and news.
- Releases: new versions of tracked agents.
- Incidents: outages from official status pages.
- Deals: new official offers.
For language models #
- llms.txt: a short map of every section of the site.
- llms-full.txt: the whole catalog in one file, one line per page.
- docs/llms.txt: every Docs page.
- Any Docs page as Markdown: add
.mdto its address, for example /docs/kits/use-a-kit.md.
A stack for a goal #
GET https://agentgid.com/app/api/builder/suggest?goal=<what you are building> returns the parts and suggested tools as JSON. To show a person the result, link to https://agentgid.com/build/?goal=<goal>.
For AI assistants explains how an assistant should use all of this and cite it.
Something wrong on this page? Tell us by e-mail, or see Contact.