PumpCrateDocsAgent access: Markdown, llms.txt and A2A
Besides the MCP server, PumpCrate offers several plain ways for agents and crawlers to read its content. All of them are open and need no key.
Every page as Markdown#
Request any page on www.pumpcrate.com with an Accept: text/markdown header and you get a clean Markdown version instead of HTML: title, description and the page content, without navigation or scripts.
curl -s -H 'accept: text/markdown' https://www.pumpcrate.com/
curl -s -H 'accept: text/markdown' https://www.pumpcrate.com/pump-fun-glossary/llms.txt#
| File | What it holds |
|---|---|
https://www.pumpcrate.com/llms.txt | A condensed summary of the main site: what PumpCrate is, prices, rules and links to the main pages and posts |
https://docs.pumpcrate.com/llms.txt | An index of every docs page with its description |
https://docs.pumpcrate.com/llms-full.txt | The full text of every docs page in one plain text file |
A2A endpoint#
PumpCrate also answers the Agent2Agent protocol. The agent card describes three skills: facts, limits and pages.
| Property | Value |
|---|---|
| Agent card | https://www.pumpcrate.com/.well-known/agent-card.json |
| Endpoint | https://www.pumpcrate.com/a2a |
| Method | JSON-RPC message/send |
| Protocol version | 0.3.0 |
curl -s -X POST https://www.pumpcrate.com/a2a \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"message/send","params":{"message":{"role":"user","kind":"message","messageId":"m1","parts":[{"kind":"text","text":"What do the packages cost?"}]}}}'Discovery files#
| Path | Purpose |
|---|---|
/.well-known/ai-catalog.json | Catalogue of the readable resources on the site |
/.well-known/mcp/server-card.json | MCP server description, tools and auth |
/.well-known/agent-card.json | A2A agent card |
/.well-known/oauth-authorization-server | OAuth authorization server metadata (RFC 8414) |
/.well-known/oauth-protected-resource | Protected resource metadata (RFC 9728) |
/auth.md | Plain text guide to agent authentication |
/sitemap.xml, /robots.txt | Standard crawl files; AI crawlers are allowed |
For one-off questions, llms.txt or a Markdown page is the lightest option. For an assistant that answers PumpCrate questions regularly, connect the MCP server.