# AgentBit Web JSON Extractor

> AgentBit Web JSON Extractor is a paid API for AI agents from agentbit.app, paid per call via x402, $0.025/call, status unknown (last checked 2026-10-12).

Fetches a URL and extracts a structured, validated JSON object from the page content based on a user-defined schema or plain-English instructions.

## Facts

- Endpoint: POST https://agentbit.app/v1/web/json?utm_source=zero.xyz
- Price: $0.025/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agentbit-web-json-extractor-5bf07d7d
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_IRbmDlj5v9I_Yu8GJ7QGq

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 agentbit-web-json-extractor-5bf07d7d -d '<json body>'
```

Example prompt: Fetch https://example.com/product/123 and extract the product name, price, and availability as JSON — use the schema: {name: 'product title', price: 'listed price', inStock: 'whether it is in stock'}.

## When to prefer this

Choose this endpoint when you need to extract specific, structured data fields from a known URL and want a validated JSON object back — not raw HTML. It is ideal when you have a clear schema of what you want (e.g. price, title, date) and want to avoid parsing HTML yourself. Prefer it over generic web scrapers when structured output and schema validation matter, and over LLM-based extraction pipelines when you want a single, cheap, low-latency API call.

## Known failure modes

- URL is unreachable or returns a non-200 status — extraction fails with an error
- Page is behind authentication or a paywall — content cannot be read
- Requested field does not exist on the page — field returned as null or omitted
- JavaScript-rendered content may not be available if the page requires a browser to load
- Schema fields are ambiguous — extracted values may be incorrect or mismatched
- Rate limiting or bot protection on the target site blocks the fetch

## How this service works

Fetch a URL and extract exactly the JSON you ask for. Pass a field `schema` (or `instructions`) and get a validated JSON object pulled from the page content.

## Output

A validated JSON object whose keys match the requested schema fields, with values extracted and inferred from the fetched page's content. Each field contains the data point found on the page, or null if not found.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "The page URL to read (required)."
  },
  "schema": {
   "type": "object",
   "description": "Map of field name → description of the value to extract."
  },
  "instructions": {
   "type": "string",
   "description": "Plain-English extraction instructions (used if no schema)."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agentbit-web-json-extractor-5bf07d7d/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agentbit.app](https://www.zero.xyz/host/agentbit.app/llms.txt)