// web tools for AI agents
Search. Fetch. Extract.
One API to search the web, fetch any page as clean markdown, extract structured data, or run deep research.
Search
1–10 credits per request
POST /v1/search
{
"query": "best restaurants in NYC",
"search_depth": "basic",
"max_results": 10
}
response
{
"query": "best restaurants in NYC",
"results": [
{
"title": "The 50 Best Restaurants in NYC Right Now",
"url": "https://www.eater.com/nyc-best-restaurants",
"content": "From fine dining to hole-in-the-wall gems...",
"description": "A curated guide to the best restaurants...",
"fetched": true,
"published_date": "2026-03-16"
}
],
"search_depth": "basic",
"topic": "general",
"elapsed_ms": 4200,
"credits_used": 3,
"credits_remaining": 997
}Fetch
1 credit per request
POST /v1/fetch
{
"url": "https://example.com"
}
response
{
"title": "Example Domain",
"url": "https://example.com",
"content": "# Example Domain\n\nThis domain is for use in
documentation examples...",
"published_time": null,
"credits_used": 1,
"credits_remaining": 999
}Extract
5 credits per request · paid plans only
POST /v1/extract
{
"url": "https://example.com",
"prompt": "Summarize this page in one sentence"
}
response
{
"content": "This domain is reserved for documentation
purposes only and should not be used in
actual operations.",
"url": "https://example.com",
"credits_used": 5,
"credits_remaining": 995,
"usage": { "input_tokens": 90, "output_tokens": 24 }
}Research
25 credits per request · paid plans only
POST /v1/research
{
"query": "How do modern LLMs handle long context?",
"max_sources": 20
}
response
{
"query": "How do modern LLMs handle long context?",
"report": "## Long Context in Modern LLMs\n\nRecent advances...",
"sources": [
{
"title": "Scaling Transformer Context Windows",
"url": "https://arxiv.org/abs/...",
"fetched": true
}
],
"sub_queries": [
"transformer context window scaling techniques",
"RoPE positional encoding extensions"
],
"credits_used": 25,
"credits_remaining": 975,
"usage": { "input_tokens": 12400, "output_tokens": 1850 }
}