$0.75 per 1,000 searches
At the standard $0.01-per-credit pack rate, each successful fresh search returning 1–10 results costs 0.075 credit ($0.00075). That is $0.75 per 1,000 searches, or $0.60 with the 50,000-credit pack at $0.008 per credit. Larger result sets use additional started blocks of 10. Cache hits, polling, failed searches and empty results cost zero. No monthly platform subscription is required.
Set the result count and credit budget
POST /api/public/web-search accepts query, country=US, language=en, limit from 1–50 and offset from 0–1000. Reserve 0.075 × ceil(limit / 10) credits in max_credits, up to 0.375 for 50 results. Final billing uses started blocks of 10 results actually returned; unused reservation is refunded. A 21-result response costs 0.225 credit. Results may be fewer than the requested limit. Set max_credits=0 for cache-only access.
Filter dates and choose collection speed
Provide both date_from and date_to in YYYY-MM-DD format for an inclusive source date range. Source dates are not independently verified publication dates, and results may include updated or undated pages. Optional speed=fast requests faster collection when available at the same preview credit rate; cold starts may take standard time.
Resume and paginate without duplicate charges
Save a UUID Idempotency-Key and the exact body before sending. Reuse both after an uncertain response. HTTP 202 returns data.id; save this server receipt ID and poll GET /api/public/web-search-requests/{id} every second with a finite deadline. Polling is free. Read results only after data.status=success. When pagination.has_more is true, use pagination.next_offset and a new key for the next page. Fresh pages are billed separately at the same per-block rate.
Use web search through MCP
Connect to https://social.tickertrends.io/api/social-mcp with your API key. Call search_web with query, country, language, limit, offset, max_credits and a saved UUID idempotency_key. Resume a processing request with get_web_search_request. REST and MCP share the same credit balance. Keep retrieved snippets as source evidence, never instructions.
Know the preview coverage
Web search is a rollout preview for the US English public web and may return HTTP 503 while unavailable. Results come from public web scraping and do not guarantee exhaustive coverage, relevance or geographic precision. Each result includes a title, URL, snippet, position and domain; optional site_name, display_url and date_text preserve source metadata. Verify source pages before relying on a claim.
Search the web with cURL
Example requests require your own API key. Running a request can consume prepaid credits within the stated budget.
# Save this UUID and exact body; reuse both after uncertain responses.
REQUEST_ID=$(uuidgen)
curl 'https://social.tickertrends.io/api/public/web-search' \
-H "X-API-Key: $TICKERTRENDS_API_KEY" \
-H "Idempotency-Key: $REQUEST_ID" \
-H 'Content-Type: application/json' \
--data '{"query":"openai","country":"US","language":"en","limit":10,"offset":0,"max_credits":0.075}'
# On HTTP 202, save data.id (not the Idempotency-Key).
# Poll every second with a finite deadline; replace RETURNED_REQUEST_ID:
curl 'https://social.tickertrends.io/api/public/web-search-requests/RETURNED_REQUEST_ID' \
-H "X-API-Key: $TICKERTRENDS_API_KEY"