# Pocket Network Web Extract (Markdown/Text)

> Pocket Network Web Extract (Markdown/Text) is a paid API for AI agents from agent.pocket.network, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-02).

Fetches a URL and returns it as clean markdown, plain text, title, links, and fetch metadata — formatted for LLM or RAG pipeline ingestion.

## Facts

- Endpoint: POST https://agent.pocket.network/v1/agentsearch-web-extract-v1?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/pocket-network-web-extract-markdown-text-c44f6a47
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_n1EvGnqwZD_0CwzDm6fn9

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 pocket-network-web-extract-markdown-text-c44f6a47 -d '<json body>'
```

Example prompt: Fetch https://en.wikipedia.org/wiki/Large_language_model and give me back the content as clean markdown, keeping headings and links, capped at 20000 characters.

## When to prefer this

Choose this endpoint when you need a URL's content cleaned and structured specifically for LLM or RAG use — it returns markdown with headings/links intact plus plain text in a single call, billed per-request in USDC with no API key required. Prefer it over generic scraping APIs when your downstream consumer is a language model or embedding pipeline and you want immediate, pay-as-you-go access without account setup.

## Known failure modes

- URL is unreachable or returns a non-200 HTTP status — error included in JSON response
- Page requires JavaScript rendering not supported by the fetcher — returns partial or empty content
- max_chars too low — content silently truncated at the specified limit
- Paywalled or login-gated pages — returns only publicly visible content (e.g. login prompt text)
- Malformed URL input — error returned in response JSON
- USDC payment failure — request rejected before fetch is attempted

## How this service works

Fetch a URL and get it back ready for an LLM or RAG pipeline: clean markdown (headings and links kept), plain text, the title, the page's links and fetch metadata. POST /v1/extract with {url} and optional formats and max_chars. JSON in, JSON out, errors included. Pay per request in USDC; no account, no API key.

## Output

A JSON object containing one or more of: 'markdown' (headings and links preserved, HTML stripped), 'text' (plain prose), 'title' (page title string), 'links' (array of URLs found on the page), and fetch metadata (status, fetch timestamp, source URL). Characters are capped at max_chars (default 50000). Errors are returned inline in the JSON response.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "The page to fetch (http or https)."
  },
  "formats": {
   "type": "array",
   "items": {
    "enum": [
     "markdown",
     "text"
    ],
    "type": "string"
   },
   "description": "Which renderings to return. Defaults to the service's own choice."
  },
  "max_chars": {
   "type": "integer",
   "maximum": 100000,
   "minimum": 1,
   "description": "Cap on returned characters (default 50000)."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/pocket-network-web-extract-markdown-text-c44f6a47/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent.pocket.network](https://www.zero.xyz/host/agent.pocket.network/llms.txt)