API reference · Context.dev

context-dev/extract-structured-data

Integrate this model through SandBase's unified API, with production-ready schemas and examples.

APIAsyncOpen model

Production endpoint

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

POSThttps://api.sandbase.ai/v1/run

Model IDcontext-dev/extract-structured-data

Input Schema

13 parameters · 2 required · 11 optional

Parameter	Type	Required	Description
`url`	string	Required	Starting website URL to crawl and extract from.
`schema`	object	Required	JSON Schema describing the object to return. Generate it from a Zod or Pydantic model, or hand-write it — Context.dev fills exactly this shape.
`maxAgeMs`	integer	Optional	Reuse a cached result younger than this many milliseconds. Default 86400000 (1 day), max 2592000000 (30 days). Set 0 to always fetch fresh. · Min: 0 · Max: 2592000000
`maxDepth`	integer	Optional	Maximum link depth from the starting URL (0 = only the starting page). Unlimited when omitted. · Min: 0 · Max: 9007199254740991
`maxPages`	integer	Optional	Maximum number of pages to analyze. Default 5, hard cap 50. Does NOT change the price — extraction is billed per call. · Min: 1 · Max: 50
`factCheck`	boolean	Optional	When true, every returned value must be grounded in text stated on the page and unsupported fields come back null/empty. When false (default), reasonable inferences are allowed while verifiable specifics stay faithful to the source.
`timeoutMS`	integer	Optional	Upstream timeout in milliseconds (max 300000). Context.dev aborts the call with a 408 when exceeded. Keep it below the endpoint's requestTimeoutMs so the provider answers before the platform's own budget expires. · Min: 1000 · Max: 300000
`waitForMs`	integer	Optional	Extra browser wait in milliseconds after page load before the content is captured (0-30000). Useful for JavaScript-heavy pages. · Min: 0 · Max: 30000
`stopAfterMs`	integer	Optional	Soft time budget for the crawl phase in milliseconds (10000-110000, default 80000). · Min: 10000 · Max: 110000
`instructions`	string	Optional	Extraction guidance: which facts to prioritize, how to interpret ambiguous fields. · Max length: 2000
`includeFrames`	boolean	Optional	Include iframe contents in the Markdown handed to the extractor. Default false.
`followSubdomains`	boolean	Optional	Follow links on subdomains of the starting URL's domain. Default false.
`settleAnimations`	boolean	Optional	Wait briefly for animations to settle before each page is read. Default false.

Output Schema

Field	Type	Description
`id`	string	Unique identifier for the generation task
`status`	string	Task status: pending, running, completed, failed, timeout
`model`	string	Model used for the generation
`outputs`	array	Array of output items
`outputs[].url`	string	URL of the generated artifact
`outputs[].content_type`	string	MIME type (e.g. image/png, video/mp4)
`error`	object \| null	Error details if failed, null on success
`error.type`	string	Machine-readable error type code
`error.message`	string	Human-readable error description

Async Workflow

This model uses asynchronous execution. Submit a request and poll for the result.

Submit — POST to /v1/run, receive an id
Poll — GET /v1/run/{id} until status is completed, failed, or timeout
Retrieve — Read outputs from the completed response

Code Examples

Ready-to-run snippets

# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
  "model": "context-dev/extract-structured-data",
  "prompt": "a beautiful sunset over mountains"
}'

# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
  -H "Authorization: Bearer YOUR_API_KEY"

← Back to Extract Structured Data