Wikexa
Structured Wikipedia + Wikidata for AI agents. Clean JSON, edge-served, free.
Fetching a Wikipedia page costs an agent roughly 15,000 tokens of HTML to recover about 500 tokens of facts. Wikexa returns the facts.
MCP endpoint
POST https://wikexa.com/mcp
JSON-RPC 2.0 — initialize, tools/list,
tools/call, ping. No API key, no auth.
Tools
| Tool | Returns |
|---|---|
lookup | Structured facts + ~200-token summary. Resolves aliases and Q-ids. |
article | Full article, sections as JSON. sections and max_chars control token cost. |
search | Ranked entity matches when the exact title is unknown. |
recent | Articles changed recently — facts newer than your training cutoff. |
Try it
curl -s https://wikexa.com/v1/lookup?entity=Tim%20Cook
curl -s -X POST https://wikexa.com/mcp \
-H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
"params":{"name":"lookup","arguments":{"entity":"Apple Inc."}}}'
Also
llms.txt · openapi.json · mcp.json · stats