Verifiable · Agent-native · MCP-ready

The verifiabledocument layerCryptographically-grounded document memory — tamper-evident bundles, SHA-256 + Merkle proofs, and free offline verification. Every document your agent touches, provably trustworthy.for AI agents.

Give your agents cryptographically-grounded document memory — tamper-evident evidence bundles, SHA-256 + Merkle proofs, and free offline verification. Every document, provably trustworthy.

from $0.018/call·verify offline · always free·MCP + x402·SHA-256 · NIST FIPS 180-4

Run a live request Get free API key

Read the docs →

Citation-grounded results·USDC pay-per-call·SHA-256 + Merkle proofs

x402 payments·For agents·Agent card ↗

High-trust documents — not open-web scraping.

POST /v1/extract → capture · analyze · prove

one call/extract

Capture

OCR · async · webhooks

Analyze ⟷ Cite

schema · Q&A · claim-check

Evidence ⟷ Verify

SHA-256 · free verify · hold

Matter corpus

search · cross-doc ask

Agent access

x402 · API keys · MCP

one call/extract

Capture

OCR · async · webhooks

Analyze ⟷ Cite

schema · Q&A · claim-check

Evidence ⟷ Verify

SHA-256 · free verify · hold

Matter corpus

search · cross-doc ask

Agent access

x402 · API keys · MCP

jsonResponse

{
  "bundle_id": "ev_abc123",
  "manifest_sha256": "a3f2c1d8e9…",
  "artifacts": {
    "markdown": { "sha256": "7b4e…" },
    "screenshot": { "sha256": "c91f…" }
  },
  "verify": "https://api.docimprint.com/v1/extract/ev_abc123/verify — free"
}

Evidence · verify · notarize

Tamper-evident evidence agents can attach to decisions.

Every extract can produce a bundle agents attach to reports and audit trails — provable truth with cryptographic hashes, free offline verify, and optional on-chain anchoring.

Evidence docs →

Evidence bundle
Markdown + screenshot + OCR + structured data in one tamper-evident package with SHA-256 per artifact.
Free verification
GET /v1/extract/{id}/verify — anyone can confirm bundle integrity, no payment required.
ZIP download
GET /v1/extract/{id}/download — hand a single archive to auditors or compliance workflows.
Line-level citations
Claim-check and structured responses include source paragraph pointers so reviewers can validate outputs.
On-chain notarize
POST /v1/extract/{id}/notarize can anchor the manifest hash on Base when notarization is enabled.

On-chain notarization anchors a cryptographic hash — it is not a legally recognized notarial act. AI outputs may contain errors; verify before relying on them professionally.

POST /v1/extract · all-in-one

POST /v1/extract collapses common document workflows.

For common document flows, agents no longer stitch screenshot + OCR + parsing + hashing by hand. Extract consolidates the sequence and returns a verifiable bundle.

Without DocImprint

bash

# Agent chains tools manually

POST /screenshot → PNG bytes
POST /ocr → flat text (tables garbled)
POST /summarize → unverified summary
# manually hash artifacts? good luck.

# 4 calls · 4 payments · no provenance

With POST /v1/extract

bash

# One DocImprint call

POST https://api.docimprint.com/v1/extract
  -H "Content-Type: application/pdf"
  -H "X-Payment: <signed>"
  --data-binary @contract.pdf

→ {
    "bundle_id": "ev_...",
    "manifest_sha256": "a3f2c1...",
    "artifacts": { "markdown", "screenshot", "ocr" },
    "structured_data": { "parties": [...] },
    "citations": [{ "paragraphs": [2] }]
  }

# 1 call · 1 payment · tamper-evident bundle

Firecrawl

Output

Markdown chunks

Proof

None

Agent pay

API keys

Verify

—

LlamaParse

Output

Structured JSON

Proof

Optional citations

Agent pay

API keys

Verify

—

DocImprint

Output

✓Evidence bundle + hashes

Proof

✓Free offline verify

Agent pay

✓x402 USDC + API keys

Verify

✓Free offline + on-chain

Based on public documentation. Firecrawl and LlamaParse are trademarks of their respective owners.

What we are not

−A Firecrawl competitor for open-web scraping
−A general-purpose web crawler or monitor

URLs are an ingestion path for hosted PDFs — the product is high-trust verifiable document memory.

x402 protocol

Request → 402 → Pay and receive bundle.

No account required. Agent sends HTTP, pays USDC on Base, gets the evidence bundle — or use API keys for account-based usage.

Agent sends a request

Standard HTTP POST request for extract

http

POST /v1/extract HTTP/1.1
Host: api.docimprint.com
Content-Type: application/json

{"source": "contract.pdf"}

Gateway returns a 402

Payment challenge with amount, asset, and network (example shows Base Sepolia)

http

HTTP/1.1 402 Payment Required
X-Payment-Required: true
X-Payment-Amount: 75000
X-Payment-Asset: 0x036CbD53...
X-Payment-Network: eip155:84532

Agent pays and gets the result

Re-sends with signed payment — receives the evidence bundle

http

POST /v1/extract HTTP/1.1
X-Payment: <eip712-signed-usdc-transfer>

→ HTTP/1.1 200 OK
   Content-Type: application/json
   { "bundle_id": "ev_...",
     "manifest_sha256": "abc123...",
     "structured_data": { "parties": [...] } }

POST /v1/extract · modes

POST /v1/extract. Verifiable capture modes.

BYOS, claim-check, invoice parse, summarize, Q&A, compare — all via POST /v1/extract on contracts, filings, and PDFs.

BYOS

$0.075

Bring-Your-Own-Schema: pass a JSON schema and get guaranteed structured output from any contract or filing.

{ bundle_id, structured_data, citations[], manifest_sha256 }

bashRequest

curl -X POST https://api.docimprint.com/v1/extract \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer dr_live_..." \
  -d '{
  "source": "https://cdn.example.com/contract.pdf",
  "schema": {
    "parties": ["string"],
    "effective_date": "string",
    "governing_law": "string"
  }}'

jsonResponse

{
  "bundle_id": "ev_...",
  "structured_data": {
    "parties": ["Acme Corp", "Beta LLC"],
    "effective_date": "2025-01-01",
    "governing_law": "Delaware"
  },
  "citations": [{ "field": "effective_date", "paragraphs": [2] }],
  "manifest_sha256": "abc123..."
}

Clean Markdown

Token-efficient, LLM-native text stripped of boilerplate.

BYOS JSON

Your schema, guaranteed shape — parties, dates, amounts, clauses.

Evidence Bundle

SHA-256 manifest + artifacts stored in cloud storage.

Citations

Line and paragraph pointers back to source passages.

bashRequest

curl -X POST https://api.docimprint.com/v1/extract \
  -H "Content-Type: application/pdf" \
  -H "X-Payment: <eip712-signed-usdc-transfer>" \
  --data-binary @contract.pdf

jsonResponse

{
  "bundle_id": "ev_abc123",
  "manifest_sha256": "a3f2c1...",
  "artifacts": {
    "markdown": { "sha256": "..." },
    "screenshot": { "sha256": "..." },
    "ocr": { "sha256": "..." }
  },
  "structured_data": { "parties": ["Acme Corp", "Beta LLC"], "effective_date": "2025-01-01" },
  "citations": [{ "field": "effective_date", "paragraphs": [2] }]
}

high-stakes documents

What will your agent build?

High-stakes tasks map to one extract surface: contracts, financial filings, research papers, invoices, regulatory docs, and insurance policies.

ContractsFinancial filingsResearch papersInvoices & receiptsRegulatory docsInsurance policies

Process a contract

Upload a PDF and get an Evidence Bundle plus BYOS extraction of parties, dates, and governing law.

Extract$0.075

Verify claims against a filing

Check assertions against a 10-K or research paper. Returns supported / contradicted / not_found with citations.

Extract$0.075

Parse an invoice

Extract line items, totals, vendor name, and dates from receipts and invoices in one call.

Extract$0.075

Extract tables from a financial PDF

Spatial OCR preserves layout — tables returned as Markdown syntax agents can reliably parse.

Extract$0.075

Summarize a research paper

Clean Markdown plus AI summary in a verifiable bundle — attach to agent reports and decisions.

Extract$0.075

Prove what the agent read

Verify bundle integrity for free. Optionally notarize the manifest hash on Base for audit trails.

VerifyFree

Discovery, verification, and status endpoints are always free — no payment required.

All use cases →Legal & compliance →Finance & audit →

A2A · MCP · OpenAPI

Agents discover and integrate automatically.

Machine-readable discovery, native MCP tools, and HTTP examples — no SDK required.

A2A↗

{
  "mcpServers": {
    "docimprint": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://api.docimprint.com/mcp",
        "--header",
        "Authorization:${DOCIMPRINT_AUTH}"
      ],
      "env": {
        "DOCIMPRINT_AUTH": "Bearer dr_live_YOUR_KEY"
      }
    }
  }
}

MCP tools call the same intelligence layer as REST. OpenAPI and SKILL.md cover HTTP agents and marketplace discovery.

extract_textPDF or image → raw text via OCR
extract_tablesPDF or image → Markdown tables
parse_invoiceInvoice/receipt → merchant, line items, totals
check_claimsDocument + claims[] → supported / contradicted / not_found
extract_structuredDocument + JSON schema → typed fields with citations
summarize_documentDocument → summary + key_points[]
extract_urlHTTPS URL → page text
summarize_urlHTTPS URL → summary + key_points[]
qa_urlHTTPS URL + question → cited answer
translate_urlHTTPS URL + target language → translated text
verify_bundlebundle_id → integrity check (returns receipt)
get_bundlebundle_id → metadata + artifacts (returns receipt)
notarize_bundleAnchor manifest hash on Base L2 (returns receipt)
verify_action_receiptrcpt_… → signature + manifest binding check
create_collectionCreate a named document corpus
list_collectionsList collections you own
add_document_to_collectionIndex a bundle into a collection (returns receipt)
search_collectionSemantic search over indexed chunks
ask_collectionCross-document Q&A with citations
get_job_statusPoll async job by job_id
get_quotaCredits used / remaining (free)

bash

# 1. Request — gateway responds with 402
curl -X POST https://api.docimprint.com/v1/extract \
  -H "Content-Type: application/pdf" \
  --data-binary @contract.pdf

# 2. Re-send with signed x402 payment
curl -X POST https://api.docimprint.com/v1/extract \
  -H "Content-Type: application/pdf" \
  -H "X-Payment: <eip712-signed-transfer>" \
  --data-binary @contract.pdf

For agents →·Quickstart →·MCP docs →

The verifiabledocument layerCryptographically-grounded document memory — tamper-evident bundles, SHA-256 + Merkle proofs, and free offline verification. Every document your agent touches, provably trustworthy.for AI agents.

What is verifiable document memory?

How do AI agents pay for DocImprint without an account?

What makes DocImprint different from web scrapers and RAG?

What is verifiable document memory?

How do AI agents pay for DocImprint without an account?

What makes DocImprint different from web scrapers and RAG?

Tamper-evident evidence agents can attach to decisions.

POST /v1/extract collapses common document workflows.

Every step is verifiable — from capture to proof.

Request → 402 → Pay and receive bundle.

POST /v1/extract. Verifiable capture modes.

BYOS

What will your agent build?

Process a contract

Verify claims against a filing

Parse an invoice

Extract tables from a financial PDF

Summarize a research paper

Prove what the agent read

Agents discover and integrate automatically.

Agent Card

OpenAPI 3.1

MCP Server

Base USDC

No surprises. Ever.

Run your first extract in 60 seconds.