# Fresh Structured Extract

> Fresh Structured Extract is a paid API for AI agents from mcp.dropenginehq.com, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-02).

Fetches a public webpage and extracts user-defined structured JSON fields from its live content.

## Facts

- Endpoint: POST https://mcp.dropenginehq.com/api/extract-structured?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/fresh-structured-extract-bc098964
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_23jusZ8LxXQJOJxmL2WsF

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability fresh-structured-extract-bc098964 -d '<json body>'
```

Example prompt: Can you fetch https://techcrunch.com/2024/01/15/openai-funding/ and extract the article title, author name, publication date, and funding amount as structured JSON?

## When to prefer this

Use this endpoint when you need to pull specific, named fields from a live public webpage and receive them as typed JSON — especially when the target page changes frequently and you need fresh data rather than cached results. Prefer this over general-purpose scrapers when you want schema-enforced output rather than raw HTML or markdown. Ideal for agents that need to feed structured data into downstream logic without manual parsing.

## Known failure modes

- URL is not publicly accessible or requires authentication — extraction fails
- Webpage content does not contain the requested fields — fields return null or are missing
- Target site blocks automated requests — may return empty or partial results
- Malformed or unreachable URL — validation error returned
- Schema field types do not match extractable content — type coercion or null returned

## How this service works

Extract fresh structured JSON from a public webpage.

## Output

Returns a structured JSON object whose keys match the schema the caller provided, with values populated from the live webpage content. Each field is typed according to the requested schema (string, number, boolean, or nullable variants). The data reflects the current state of the page at call time.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "format": "uri",
   "maxLength": 4096
  },
  "schema": {
   "type": "object",
   "propertyNames": {
    "type": "string"
   },
   "additionalProperties": {
    "enum": [
     "string",
     "number",
     "boolean",
     "string|null",
     "number|null",
     "boolean|null"
    ],
    "type": "string"
   }
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/fresh-structured-extract-bc098964/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from mcp.dropenginehq.com](https://www.zero.xyz/host/mcp.dropenginehq.com/llms.txt)