---
title: "Extract readable text from supplied HTML, with an explicit generic fallback and exact input provenance. Does not fetch the source URL."
method: POST
path: "/api/v1/capabilities/web.extract_evidence/run"
---

# Extract readable text from supplied HTML, with an explicit generic fallback and exact input provenance. Does not fetch the source URL.

`POST /api/v1/capabilities/web.extract_evidence/run`

## Request

- Base URL: `https://api.daystruct.ai`
- URL: `https://api.daystruct.ai/api/v1/capabilities/web.extract_evidence/run`
- Auth: HTTP bearer

data.status completed means extraction ran; inspect result.status (extracted or empty), result.extractor.selected and warnings. Empty is not proof there is no useful content. Rejected inputs return 400; extraction or worker failure returns 503.

## Request body

- object
  - `html` string, required
  - `sourceUrl` string, required: HTTP(S) URL without credentials; caller asserted, never fetched
  - `capturedAt` string, date-time, required: Valid UTC timestamp with optional three-digit milliseconds
  - `maxInputBytes` integer

## Response `200`

Execution completed; inspect data.result.status

## Other responses

- `400`: Rejected input
- `401`: Invalid or missing key
- `403`: Insufficient scope
- `429`: Capacity or rate limit
- `503`: Execution failed or unavailable

---

[API](https://skmtc.dev/daystruct/apis/daystruct-api.md) · [All operations](https://skmtc.dev/daystruct/apis/daystruct-api/llms.txt) · [OpenAPI document](https://skmtc.dev/daystruct/apis/daystruct-api/revisions/8cd3c7f60e18?raw)