Every source in the catalog is below, grouped by category. Each links to a reference page generated from the published schema, so what you read is what the API actually accepts and returns — there is no hand-maintained copy to drift.
204 sources across 24 categories, all behind one key, one host and one response shape.
What every source has in common
Before the list, the parts that never change:
| How it appears | One tool per source, generated from the catalog |
| Auth | The connection your client made — see authentication |
| What comes back | structuredContent matching the tool's output schema |
| Failures | A result with isError: true and a readable reason — see errors |
So the difference between any two sources is the input schema and the shape of the result. Everything else the protocol handles.
Reading a reference page
Each page follows the same order, and each section answers a specific question:
- Overview — what it is for, and when it is the wrong tool.
- Tool — the tool name and a worked call.
- Parameters — name, type, required or optional, constraints, and accepted values where the set is fixed. Premium parameters are marked.
- Response — a real sample response, not a schema sketch.
- Response fields — every field with its dot-path, type, an example value and a description.
- Errors — what it rejects and why.
- Use cases — what people actually build with it.
Credit cost
Most sources cost one credit per call. The ones that do more work — rendering, OCR, inference, multi-source lookups — cost more, and their cost is shown on the card below and again at the top of their reference page.
Failed calls are not billed. A credit is taken only when a source actually answers, so a bad parameter, a rejected credential and a blocked call all cost nothing — the request never reaches the source that would charge for it.
The case that is billed and surprises people is a successful call that found nothing. A lookup returning an empty result did the work you asked for, and it costs its credit. An empty answer is an answer — see errors.
The catalog
Finance 50
Domain Data 18
Data Lookup 16
Networking 14
Geography 12
Reference Data 11
Data Conversion 10
Weather 9
Data Validation 9
Math/Calculations 8
Transportation 6
Calendar 5
Data Processing 5
AI/Computer Vision 4
Data Scraping 4
Science 4
Astrology 4
Food 3
Lifestyle 3
Parsers 3
Health/Wellness 2
Entertainment 2
Security 1
News 1
Next
Pick anything above and its reference page has a working call you can paste. If you have not made one yet, quickstart is the fastest route there.
Terms used across these pages are defined in the glossary.