# Agent Rescue Desk — PDF Citation Pack

> Agent Rescue Desk — PDF Citation Pack is a paid API for AI agents from agent-rescue-desk.netlify.app, paid per call via x402, $0.25/call, status unknown (last checked 2026-09-14).

Extracts structured, SHA-256-addressable citation blocks from text-layer PDFs so AI agents can cite sources with verifiable evidence addresses

## Facts

- Endpoint: POST https://agent-rescue-desk.netlify.app/api/agent402/pdf-citation-pack
- Price: $0.25/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agent-rescue-desk-pdf-citation-pack-4a484854
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_VighVQHbfuFv83w-k0peW

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 agent-rescue-desk-pdf-citation-pack-4a484854 -d '<json body>'
```

Example prompt: Extract all citable text blocks from this research paper PDF so I can reference specific passages with verifiable SHA-256 addresses — here are the base64-encoded bytes for the document with id 'paper-2024-q3'.

## When to prefer this

Choose this endpoint when an AI agent needs to extract text from text-layer PDFs and produce verifiable, citable output with SHA-256-addressable block IDs. It is ideal for RAG pipelines, legal or academic document workflows, and any agent task requiring evidence traceability. It is not suitable for scanned/image PDFs (no OCR), redaction, accessibility certification, or layout-fidelity work. Prefer this over generic PDF parsers when citation integrity and agent-usable block IDs are the priority.

## Known failure modes

- Batch exceeds 2,000,000 decoded bytes — request rejected
- More than 40 pages submitted across all documents — request rejected
- PDF is image-only (no text layer) — this is not OCR, extraction will be empty or minimal
- Unusual filter chains or high-expansion streams rejected by budget limits
- Oversized string tokens rejected
- Submission of passwords, regulated data, payment-card data, health data, or private keys — rejected by policy
- Invalid or malformed base64 encoding — parse error
- minLength of 40 characters not met in query field — validation error

## How this service works

Fixed-scope diagnosis and repair for one broken AI agent workflow or MCP integration. $1,000. Decisive outcome in 12 hours.

## Output

A JSON response containing per-document citation blocks, each with a unique block ID (e.g. p1-b0001), page number, extracted text, bounding box coordinates, and a SHA-256 hash of the document. Also includes a markdown summary per document, page count, character count, processing time, quality warnings, and a nextStep instruction for using the SHA-256 plus block ID as a verifiable evidence address when citing.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "query": {
   "type": "string",
   "maxLength": 3000000,
   "minLength": 40,
   "description": "A JSON string containing 1-5 authorized, non-sensitive text-layer PDFs as raw padded base64. Each document needs a unique URL-safe id. The decoded batch must be at most 2,000,000 bytes and 40 pages. Server-parsed streams need direct numeric lengths and must remain within the published filter, expansion, and token budgets. Do not submit passwords, regulated data, payment-card data, health data, private keys, or unrelated customer records."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "ok": true,
  "batch": {
   "pages": 1,
   "documents": 1,
   "characters": 19,
   "decodedBytes": 594,
   "citationBlocks": 1,
   "processingMilliseconds": 12
  },
  "limits": "Bounded text-layer extraction with direct numeric stream lengths and strict filter, expansion, and string-token budgets. Unusual filter chains, high-expansion streams, and oversized string tokens are rejected. This is not OCR, redaction, legal review, accessibility certification, or layout-fidelity certification.",
  "status": "complete",
  "privacy": "Submitted PDF bytes were processed transiently and were not stored or logged by the product. The response intentionally contains the requested extracted source text.",
  "service": "Agent Rescue Desk Agent-Citable PDF Extraction Pack",
  "version": "2026-07-28",
  "nextStep": {
   "instruction": "Use each SHA-256 plus page/block id as the evidence address when an agent cites the extracted text.",
   "privateKitUrl": "https://agent402.app/agents/6c54e490-72fc-47ba-bad1-3ac47b1b464a/private-pdf-citation-kit/invoke",
   "privateKitPriceUsd": 39
  },
  "documents": [
   {
    "id": "synthetic-example",
    "blocks": [
     {
      "id": "p1-b0001",
      "page": 1,
      "text": "Hello citation pack",
      "bounds": [
       0.117647,
       0.075758,
       0.159098,
       0.015152
      ]
     }
    ],
    "sha256": "69860fd72f0e874f332d71e832088707cee3c8705860b73d12681293e0b11f2d",
    "markdown": "# synthetic\\-example\n\n- SHA\\-256: `69860fd72f0e874f332d71e832088707cee3c8705860b73d12681293e0b11f2d`\n- Pages: 1\n- Extracted characters: 19\n\n## Page 1\n\n- [p1-b0001] Hello citation pack",
    "metadata": {
     "PDFFormatVersion": "1.4"
    },
    "pageCount": 1,
    "characters": 19,
    "qualityWarnings": []
   }
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agent-rescue-desk-pdf-citation-pack-4a484854/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agent-rescue-desk.netlify.app](https://www.zero.xyz/host/agent-rescue-desk.netlify.app/llms.txt)