Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/pratyay360/searchapi/llms.txt

Use this file to discover all available pages before exploring further.

SearchAPI exposes a set of HTTP endpoints built on FastAPI, each designed for a specific content type. Every endpoint accepts query parameters in the URL, requires no authentication, and returns a JSON array of URLs (or DOIs for the paper search endpoint). The service is self-hosted — deploy it locally or on any cloud provider and call it at http://localhost:8000 by default.

Base URL

http://localhost:8000
If you have deployed SearchAPI to a remote server or a cloud platform, replace localhost:8000 with your host and port. See the deployment guide for cloud and container options.

Request format

All endpoints use GET. Parameters are passed as URL query string key-value pairs — there is no request body.
GET /search/?query=machine+learning&limit=10

Authentication

No authentication is required. All endpoints are open by default. CORS is enabled for all origins, so you can call the API from any frontend, script, or agent.

Response format

All endpoints return application/json. Successful responses return HTTP 200 with a JSON array.
[
  "https://example.com/page-one",
  "https://example.com/page-two"
]
The paper search endpoint (/search/paper) returns an array of DOI strings instead of URLs:
[
  "10.1145/3495243.3560527",
  "10.1038/s41586-021-03819-2"
]

Error responses

StatusWhen it occurs
400Bad request or search engine error on /search/, /search/engine, /search/paper
500Internal error on /searchpdfs/, /search/specific/, /repositories/, /wiki/, /books/, /news/, /useragent/
Error response shape:
{ "error": "message describing what went wrong" }

MCP server

SearchAPI ships with a built-in MCP server mounted at /mcp. All endpoints are available as MCP tools, letting AI assistants such as Claude call them directly without writing code. See the MCP server guide for setup instructions.

All endpoints

MethodPathDescription
GET/search/General web search across multiple engines
GET/search/engineSearch using a specific engine (Bing, Google, DuckDuckGo, etc.)
GET/search/paperAcademic paper search via Crossref, returns DOIs
GET/books/Book search, returns URLs
GET/news/News article search, returns URLs
GET/searchpdfs/Search for PDF documents, returns direct PDF URLs
GET/search/specific/Search for files by filetype (pdf, docx, pptx, etc.)
GET/repositories/Search GitHub and GitLab repositories
GET/wiki/Search across Wikipedia and Wikimedia sites
GET/Service status and MCP server path
GET/healthHealth check endpoint
GET/useragent/Random user agent string

Build docs developers (and LLMs) love