> ## Documentation Index > Fetch the complete documentation index at: https://docs.seltz.ai/llms.txt > Use this file to discover all available pages before exploring further. # Fetch ## OpenAPI ````yaml api-reference/openapi.json POST /v1/fetch openapi: 3.1.0 info: title: Seltz API description: >- REST API for the Seltz platform: context retrieval (`/v1/search`), RAG answers (`/v1/answer`), monitors (`/v1/monitors`), and page fetching (`/v1/fetch`). contact: name: Seltz API Support url: https://seltz.ai/contact license: name: Seltz Terms of Use url: https://seltz.ai/terms version: 1.8.0 servers: - url: https://api.seltz.ai description: Seltz API security: [] tags: - name: search description: Search operations - name: answer description: Answer operations - name: monitors description: Monitor configuration - name: records description: The delivered records - name: runs description: Run history and per-request outcomes - name: agent description: 'Agent runs: create, poll, list, cancel' - name: fetch description: Fetch operations paths: /v1/fetch: post: tags: - fetch summary: Fetch URLs as LLM-ready content operationId: fetch parameters: - name: x-api-key in: header description: 'Seltz API key: https://console.seltz.ai/api-keys' required: true schema: type: string requestBody: content: application/json: schema: $ref: '#/components/schemas/FetchRequest' example: urls: - https://example.com/ required: true responses: '200': description: >- One result per requested URL. A failure to fetch a page is still a 200, with that result's `status = "error"`. content: application/json: schema: $ref: '#/components/schemas/FetchResponse' '400': description: >- Malformed body, an out-of-bounds or duplicated `urls` entry, or an unavailable `formats` or `tier` value. Unrecognized fields are ignored. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '401': description: Invalid or missing API key. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '402': description: Insufficient credits. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '429': description: Rate limited. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' '500': description: Unexpected server error. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' components: schemas: FetchRequest: type: object description: Fetch request. properties: api_key: type: - string - 'null' description: >- API key to access the service. Either this or the `x-api-key` header must be supplied on the HTTP surface; the header takes precedence and is the documented path. On gRPC this field is the only carrier. default: null formats: type: array items: type: string description: >- Representations to return, as documented format names. Omitted, or sent empty, means `\["markdown"\]` -- a repeated field carries no presence, so the two are the same request and neither means "no formats". Documented names: "markdown" -- the main content as Markdown, boilerplate removed. An unrecognized name is rejected. Silently dropping it would return a body missing the representation the caller asked for, with nothing to signal why. default: [] tier: type: - string - 'null' description: >- The service tier, which selects the price. Send `"pro"`. Unset resolves to `"pro"`. An unrecognized value is rejected rather than defaulted: the value selects a price, and quietly billing a tier the caller did not name is worse than refusing. default: null timeout_ms: type: - integer - 'null' format: int32 description: >- Wall-clock budget for one URL, in milliseconds. Applies per URL, not to the batch: URLs are fetched concurrently, so a batch takes roughly as long as its slowest entry rather than the sum. A URL that exceeds the budget gets `error.code = "timeout"`; the others in the same request are unaffected. Unset means the service default. Values are clamped to the service maximum. default: null minimum: 0 urls: type: array items: type: string description: >- The URLs to fetch. At least one, at most 20. Each must be an absolute `http` or `https` URL of at most 2048 bytes. An empty list, more than 20 entries, a duplicate entry, or an over-long entry is rejected before any billing: duplicates because `FetchResult.requested_url` is the correlation key and a repeated key is ambiguous. A URL that parses but cannot be fetched -- a host that does not resolve, an origin that refuses, a document that is not an HTML page -- is reported inside a `200` as that URL's error result, not as a rejection. default: [] FetchResponse: type: object description: Fetch response -- one result per requested URL. properties: results: type: array items: $ref: '#/components/schemas/FetchResult' description: >- One entry per entry in `FetchRequest.urls`, successful or not, in the order the URLs were requested. Correlate on `FetchResult.requested_url` rather than on position. That is the key a duplicate URL would make ambiguous, which is why the request rejects one. default: [] ErrorEnvelope: type: object description: 'The `{"error": {...}}` envelope a non-200 carries.' required: - error properties: error: $ref: '#/components/schemas/EnvelopeError' description: The rejection. FetchResult: type: object description: The outcome for one URL. required: - status properties: content_type: type: - string - 'null' description: |- The origin's declared media type for the main document, without parameters. Unset when the origin declared none. default: null error: oneOf: - type: 'null' - $ref: '#/components/schemas/FetchError' description: >- Why the fetch failed. Set when `status` is the error status, unset otherwise. default: null fetched_at: oneOf: - type: 'null' - $ref: '#/components/schemas/google.protobuf.Timestamp' description: When the content was retrieved from the origin. default: null final_url: type: - string - 'null' description: >- The URL the content actually came from, after HTTP redirects and any client-side navigation. Equal to `requested_url` when nothing redirected. default: null http_status_code: type: - integer - 'null' format: int32 description: >- The origin's HTTP status code for the main document. Unset when no network response was observed for the navigation. default: null minimum: 0 markdown: type: - string - 'null' description: >- The page's main content as Markdown, with navigation, footers, and ads removed. default: null requested_url: type: string description: >- The URL this answers, echoed verbatim from the request -- byte for byte, never normalized, and never the post-redirect URL. This is the correlation key. Where redirects landed is `final_url`. default: '' status: oneOf: - $ref: '#/components/schemas/FetchStatus' description: >- Whether this result carries content. Branch on this field, never on whether a given content field is present: a format the page could not produce is unset on an otherwise successful result, so "markdown is absent" does not mean "the fetch failed". A failed fetch is still HTTP 200, with the error status on the result. default: ok EnvelopeError: type: object description: |- The failure inside a non-200 envelope. Its `code` is `SCREAMING` and the set is closed, because it is a pure function of the gRPC code. required: - code - message properties: code: type: string description: A stable name for the rejection. message: type: string description: What went wrong, in prose. Never an internal service name. FetchError: type: object description: Why one URL failed. properties: code: type: string description: >- A stable, machine-readable reason. A client must treat an unrecognized code as a generic failure rather than rejecting the result. Documented codes: "invalid_url" -- not a parseable absolute URL. "unsupported_scheme" -- parseable, but not `http` or `https`. "url_not_accessible" -- DNS, connection, TLS, or navigation failure reaching the origin. "timeout" -- the fetch did not finish within `timeout_ms`. "unsupported_content_type" -- the document is not an HTML page. PDFs, images, archives, and other binaries are out of scope. Refused either before the fetch, from the URL, or after it, from the media type the origin declared; the caller sees one code either way, because from their side it is one fact. "extraction_failed" -- the page was fetched, but no requested format could be produced from it. "upstream_error" -- the fetch path itself failed. Ours, not the URL's. default: '' message: type: string description: >- A human-readable explanation. Always set when this message is present: `FetchError` itself is optional on the result, so an absent error is absent whole, and there is no state where a failure arrives without a reason. For operators and logs. Never parse it -- branch on `code`. The wording of any given message may change at any time. default: '' google.protobuf.Timestamp: type: string description: A timestamp in RFC 3339 format examples: - '2025-04-11T12:00:00Z' - '2025-04-11T12:00:00.123456789Z' FetchStatus: type: string description: Whether a `FetchResult` carries content. enum: - ok - error ````