Endpoints

Initial API endpoint shape for search and scrape workflows.

Endpoint names may evolve with the public API surface. Keep client wrappers small so changes are easy to absorb.

Search

POST /v1/search

Use for natural language questions that need cited answers from the live web.

curl --fail-with-body https://api-onchain.optimai.network/v1/search \
  -H "X-API-Key: $OPTIMAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What is the current state of AI search infrastructure?",
    "limit": 5
  }'

Expected response fields:

Field	Meaning
`answer`	Cited answer text.
`sources`	Source list used by the answer.
`metadata`	Request timing, tracking, and node context when available.

Scrape

POST /v1/scrape

Use for extracting one known public URL.

curl --fail-with-body https://api-onchain.optimai.network/v1/scrape \
  -H "X-API-Key: $OPTIMAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com"
  }'

Expected response fields:

Field	Meaning
`content`	Clean text extracted from the page.
`metadata`	Fetch timing, final URL, and retrieval context when available.

Planned

POST /v1/url-seeding
POST /v1/crawl

URL Seeding and Crawl are planned surfaces. Treat them as roadmap placeholders until the API is confirmed.