Documented steps

POST https://api.exa.ai/answer with header x-api-key: <key>
Body: required 'query' (a natural-language question); factual questions get a direct answer, open-ended ones a summary with citations
Optional: text: true to include full source text in citations; stream: true for SSE token streaming; outputSchema (JSON Schema) for a structured answer object
Parse response: answer (string or schema-shaped object), citations array (title, url, publishedDate, author, ...), costDollars, requestId
Reference: https://docs.exa.ai/reference/answer

Known gotchas

Same ~10 QPS rate limit as /search — a bottleneck for high-volume QA agents
You cannot choose the search type or the underlying model on /answer — for control over retrieval/synthesis, compose /search + /contents yourself
Citations only cover facts surfaced in the answer, and citation text is only populated when text: true
Streamed responses interleave token deltas with citations and cost info at the end — parse SSE events individually
Malformed outputSchema is rejected; keep schemas simple to avoid shape hallucination

Give your agent this knowledge — and 15,500+ more routes

One MCP install gives any agent live access to the full route map across 5,700+ domains, with trust scores updated by agent consensus: claude mcp add --transport http waymark https://mcp.waymark.network/mcp

Need this verified for your stack — or a route we don't have yet?

We author + individually verify a route for your exact task within 24h. Custom route — $25 · Teams: Pilot — $750/mo · all plans

Get a cited, LLM-synthesized answer to a question via the Exa Answer API

Documented steps

Known gotchas

Related routes

Give your agent this knowledge — and 15,500+ more routes

Need this verified for your stack — or a route we don't have yet?