SynthLink Docs

A read-only JSON API that delivers collected public documents enriched with LLM-generated summaries, keywords, and categories. Start with Getting Started or jump directly to the Playground.

Overview. If you're new here, start with Getting Started to request your API key and send your first request. Once you're up and running, Concepts explains how the crawler, database, and LLM enrichment pipeline work together.

Endpoints. SynthLink exposes four read-only endpoints. GET /sources returns the catalog of available sources, update intervals, and content_source values (ingestion methods). GET /documents returns the latest collected documents from public sources. GET /insights returns LLM-generated insights — summary, keywords, and categories — for each document. If you need both in one call, GET /combined merges them into a single payload.

Reference. All requests require an API key — see Authentication for how to use it and what to do when it expires. Rate Limits covers the per-minute and per-day caps and how to handle 429 responses. For the crawler pipeline and data normalization flow, see Crawler. To understand the insight layer, see Insights. For the full list of data sources and their update intervals, see Sources. When something goes wrong, Errors lists every error code and its recommended handling. Past changes are tracked in the Changelog.

Guides. To understand when data becomes available, start with Data Availability. For practical usage patterns, see Querying Data, Integration Patterns, and Use Cases.

Operations. For pipeline health and operational signals, see Status and Monitoring. For boundaries and constraints, review Limitations. If you need a new source added, submit Request a Source.