# Gold Source Indexing Agent

> Gold Source Indexing Agent is a paid API for AI agents from agentx402pay.com, paid per call via x402, $0.05/call, status unknown (last checked 2026-10-20).

Indexes a document (text or URL) into structured sections and blocks with exact UTF-16 character spans, section paths, SHA-256 fingerprints, and coverage checks — without rewriting the content.

## Facts

- Endpoint: POST https://agentx402pay.com/api/gold-index?utm_source=zero.xyz
- Price: $0.05/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-20
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gold-source-indexing-agent-c7a1b6c2
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_LIiCynoN8YgWsPUvNMY1f

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 gold-source-indexing-agent-c7a1b6c2 -d '<json body>'
```

Example prompt: Index this document text for me — I need the exact UTF-16 character spans, section paths, and SHA-256 fingerprints for every block: [paste document text here]. Label it 'Q3 Earnings Report'.

## When to prefer this

Choose this endpoint when you need a precise, non-destructive structural index of a document — preserving original text exactly while producing machine-readable section boundaries (UTF-16 spans), hierarchical section paths, and cryptographic fingerprints. It is especially suited for RAG pipelines needing exact passage locations, change-detection workflows requiring block fingerprints, and any downstream system that must cite or retrieve specific document regions. Prefer it over generic text splitters when you need coverage guarantees, reproducible fingerprints, or support for both URL-fetched HTML/PDF and raw text input in one unified API.

## Known failure modes

- Text shorter than 50 characters or longer than 1,000,000 characters returns a validation error
- Both 'text' and 'url' supplied simultaneously triggers a 400 bad request
- Inaccessible or non-public URLs (paywalled, auth-required) cause fetch failures
- PDFs exceeding 50 pages are truncated to the first 50 pages without error
- Malformed or non-UTF-8 encoded documents may produce incorrect span calculations
- Payment failure via x402 protocol blocks the request before processing

## How this service works

Index a document without rewriting it. POST JSON {"text": "..."} (50 to 1,000,000 chars). Each section and block gets an exact [start, end) UTF-16 span, its section path and a SHA-256 fingerprint, plus a coverage check. Or {"url": "..."} for HTML or PDF.

## Output

A structured index of the document containing every detected section and block, each annotated with an exact [start, end) UTF-16 character span, a hierarchical section path, a SHA-256 fingerprint, and an overall coverage check confirming all characters are accounted for. Input can be raw text (50–1,000,000 chars) or a public URL pointing to HTML, plain text, or a PDF (up to 50 pages).

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "url": {
   "type": "string",
   "description": "Public http(s) HTML, plain-text or PDF URL (first 50 PDF pages)."
  },
  "text": {
   "type": "string",
   "description": "Document text, 50 to 1,000,000 characters. Send text or url, not both."
  },
  "filename": {
   "type": "string",
   "description": "Optional label for the document"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gold-source-indexing-agent-c7a1b6c2/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agentx402pay.com](https://www.zero.xyz/host/agentx402pay.com/llms.txt)