Getting started
Build with Monocrawl
One key and one response contract across social, search, review, commerce and developer data. Start here, then open detailed reference only when you need it.
First request
A real call is one curl
{
"success": true,
"platform": "github",
"endpoint": "/v1/github/profile",
"data": {
"handle": "torvalds",
"followers": 234567,
…
},
"credits_used": 1,
"credits_remaining": 99,
"request_id": "req_4f2b8c1d09ae37b562",
"cached": false
}$ curl "https://www.monocrawl.com/v1/github/profile?handle=torvalds" \
-H "x-api-key: mn_your_key_here"About this example
Put your key in the header and choose a platform operation. The response carries the data, the exact charge, the remaining balance and a request id for support.
This example uses github/profile. See the response contract for every field.
Go deeper
Open the right reference when you need it
Quickstart
Create an account, collect the key and make a real request.
Browse endpoints
Search every wired operation and open its focused reference page.
Use with an AI agent
Give your coding agent access to public web data.
Choose an endpoint
Start with the job you need to do, not a platform catalogue.
Build a workflow
Pasteable multi-step examples with costs read from the registry.
Use Conversations
Search social mentions, understand optional complaint labels and export the evidence.
AI & MCP
Connect your agent or build your own
Monocrawl MCP server
Connect an MCP client, discover endpoints and retrieve data with credit controls.
Set up your AI agent
Connect your app, verify access and make your first bounded data request.
LangChain
Build an agent with selected Monocrawl tools and explicit credit limits.
Vercel AI SDK
Add data retrieval to tool calling while keeping response fields and source evidence.
Platform coverage
Popular data platforms
74 external data platforms have recorded successful requests through currently eligible providers. That evidence is historical, not a current health check. These are ten commonly used starting points.
What you get
The whole contract, at a glance
Authentication
x-api-key: mn_…One account key works across the catalogue.
Request path
GET /v1/{platform}/{endpoint}The same predictable URL shape everywhere.
Response envelope
success · data · request_idOne parser handles every platform response.
Billing receipt
Charge reported per requestCheck the receipt; unresolved accounting is marked explicitly.
Why Monocrawl
Why use a unified API?
Building against dozens of sources directly means separate authentication, pagination, rate limits, error vocabularies and response formats. Monocrawl collapses those differences behind one interface.
Platform-specific fields stay in data, while operational fields stay consistent. That means the same retry, billing, cache and error-handling code works across the catalogue.
Frequently asked questions
The short answers
Where can I check the registered endpoints?
The free GET /v1/utility/endpoints response is authoritative. It contains active registry operations, their list prices and parameter schemas. Registration does not guarantee current upstream availability.
What happens when an upstream fails?
Production returns a typed error and never silently substitutes sample data. Confirmed failed collection is uncharged or refunded; an unresolved accounting outcome can report credits_used:null pending reconciliation. Read the error reference.
Can I build against something that is not wired yet?
Yes—call the same route under /sandbox. Samples are explicitly marked synthetic: true, cost nothing and never appear on /v1.
Where should I start?
Use the Quickstart for your first request, then search the endpoint directory when you know what data you need.