Search bundle — Tavily + Exa on search.gedx402.com
Shard: search · Origin: https://search.gedx402.com
Operator keys: TAVILY_API_KEY, EXA_API_KEY in Cloudflare Secrets Store (npm run ged:vault -- --push --sync-workers --from-env …). Customers pay x402 USDC — no Tavily or Exa account required.
Customer wires: run npm run pricing:report in ged-x402 (or read worst_case_price_usd from GET /v1/models?bundle=search). Do not trust static markdown prices.
AM winner (Agentic.Market)
POST https://search.gedx402.com/v1/search
Content-Type: application/json
X-PAYMENT: …
{ "query": "latest quantum computing news", "max_results": 5 }Slim Tavily-shaped body → Tavily Search API → { results: [{ rank, title, url, snippet, content }], count, answer? }.
Full Tavily API (x402 proxy)
All routes mirror Tavily REST API. Pass Tavily-native JSON bodies; responses include provider: "tavily".
| Route | Tavily endpoint | Credits (operator) |
|---|---|---|
POST /v1/tavily/search | /search | 1 basic / 2 advanced |
POST /v1/tavily/extract | /extract | ~1 per URL |
POST /v1/tavily/map | /map | 1+ |
POST /v1/tavily/crawl | /crawl | 2–20 |
POST /v1/tavily/research | /research | 15–30 |
GET /v1/tavily/research/:request_id | poll status | free |
GET /v1/tavily/usage | operator usage | free |
POST /v1/search/outcome/search-lite uses the same Tavily backend as POST /v1/search.
Exa neural search (x402 proxy)
Slim GED bodies → Exa API. Responses include provider: "exa".
| Route | Exa endpoint | Customer wire |
|---|---|---|
POST /v1/exa/search | /search | pricing:report |
POST /v1/exa/contents | /contents | scales with URL count (max 10) — pricing:report |
POST https://search.gedx402.com/v1/exa/search
Content-Type: application/json
X-PAYMENT: …
{ "query": "x402 agent marketplace APIs", "num_results": 5, "type": "auto" }POST https://search.gedx402.com/v1/exa/contents
Content-Type: application/json
X-PAYMENT: …
{ "urls": ["https://example.com"], "max_characters": 2000 }Discovery
GET https://search.gedx402.com/v1/models?bundle=search- Hub:
GET https://gedx402.com/v1/models?bundle=winnersincludesPOST /v1/search - Pricing policy: PRICING-COGS.md