{"uid":"cap_VighVQHbfuFv83w-k0peW","slug":"agent-rescue-desk-pdf-citation-pack-4a484854","name":"Agent Rescue Desk — PDF Citation Pack","description":"Fixed-scope diagnosis and repair for one broken AI agent workflow or MCP integration. $1,000. Decisive outcome in 12 hours.","url":"https://agent-rescue-desk.netlify.app/api/agent402/pdf-citation-pack","method":"POST","headers":{},"bodySchema":{"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."}}},"responseSchema":{"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":[]}]}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.25","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.25/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.25","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.25","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_D7vqGG1ozvOh4zLIu6S4L","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.25","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Extracts structured, SHA-256-addressable citation blocks from text-layer PDFs so AI agents can cite sources with verifiable evidence addresses","exampleAgentPrompt":"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'.","exampleUseCases":[{"title":"Research paper citation for RAG pipeline","prompt":"I have a PDF of a scientific paper I want my agent to cite accurately — can you extract all the text blocks with page and block IDs plus a SHA-256 hash so I can use them as verifiable evidence addresses? Here's the base64-encoded PDF with id 'climate-study-2024'."},{"title":"Legal document evidence extraction","prompt":"Extract the text from this contract PDF into structured citation blocks with SHA-256 verification so my agent can reference specific clauses by page and block ID. The document id is 'contract-nda-v2' and here's the base64 content."},{"title":"Batch academic source indexing","prompt":"I have three academic PDFs I need my agent to be able to cite — please extract all the text blocks with bounding coordinates and SHA-256 hashes for each so I can build an evidence index. The ids are 'source-a', 'source-b', and 'source-c', and here are their base64 encodings."}],"resultDescription":"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.","failureModes":["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"],"whenToPreferThis":"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.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T06:36:20.196Z","isFirstParty":false}