DocImprint Document Intelligence
OfficialProvides 10 CrewAI tools for document processing, including research, extraction, Q&A, verification, and notarization, enabling multi-agent workflows with provenance tracking.
Table of contents
Related MCP server: Invoice Parser MCP
At a glance
DocImprint turns any PDF or URL into a tamper-evident evidence bundle — structured data, AI-cited answers, and a cryptographic proof your agents can verify independently.
Input | PDF or URL |
Output |
|
Verify |
|
Choose your integration
Path | Best for | Start here |
TypeScript SDK | Node agents, typed apps | |
Python SDK | CrewAI, scripts | |
MCP server | Claude, Cursor | |
x402 USDC | Autonomous agents, no account |
See what you get
Every response includes a verifiable bundle ID, manifest hash, and cited answers tied to exact source quotes.
{
"bundle_id": "ev_01jqv8k3m2x",
"manifest_sha256": "a3f2c1d8e9b0476f8a2b1c0d9e8f7a6b5c4d3e2f1a0b9c8d7e6f5a4b3c2d1e0f",
"signature": {
"signature": "0x8f4e2a1b9c3d5e7f0a2b4c6d8e0f1a3b5c7d9e1f3a5b7c9d1e3f5a7b9c1d3e5f71c",
"signer_address": "0x742d35Cc6634C0532925a3b844Bc9e7595f0EbE",
"algorithm": "secp256k1-eip191"
},
"answer_cited": {
"value": "Section 4.2 does not permit unilateral termination without 90 days written notice.",
"citations": [
{
"quote": "Neither party may terminate this Agreement unilaterally except upon ninety (90) days prior written notice to the other party.",
"paragraphs": [42],
"page": 4,
"confidence": "high"
}
]
},
"artifacts": {
"markdown": { "sha256": "b4e5f6a79876543210fedcba9876543210fedcba9876543210fedcba9876543210ab" },
"manifest": { "sha256": "a3f2c1d8e9b0476f8a2b1c0d9e8f7a6b5c4d3e2f1a0b9c8d7e6f5a4b3c2d1e0f" }
}
}How it works
flowchart LR
source[PDF or URL] --> extract[extract / qa / checkClaims]
extract --> bundle[Evidence bundle ev_...]
bundle --> artifacts[Artifacts + manifest SHA-256]
bundle --> citations[Cited answers with quotes]
artifacts --> verify[verify - free]
verify --> notarize[notarize on Base L2 - optional]Why not plain OCR or RAG?
Plain OCR | Vector RAG | DocImprint | |
Exact source quotes | No | Approximate | Yes, with paragraph refs |
Tamper evidence | No | No | SHA-256 manifest + EIP-191 sig |
Independent verify | No | No |
|
On-chain attestation | No | No | EAS on Base L2 |
Agent audit trail | No | No | Provenance + handoff (Python) |
Use cases
Scenario | Method | Result |
Contract Q&A |
| Cited answer with confidence score |
Claim verification |
| Per-claim verdict: |
Invoice intake |
| Structured merchant, line items, totals — all cited |
Compliance archive |
| Immutable on-chain record, deletion blocked |
Install
npm install docimprint # TypeScript / Node.jspip install docimprint # Python REST client
pip install "docimprint[crewai]" # + 10 CrewAI tools & ProvenanceTrackerPython source lives in python/. PyPI: docimprint.
Quick start
Get an API key at docimprint.com — free tier available, no credit card required.
TypeScript
import { DocImprintClient } from 'docimprint'
const client = new DocImprintClient({ apiKey: 'dr_live_...' })
const result = await client.extract({
source: 'https://example.com/contract.pdf',
include: ['markdown', 'summary'],
})
console.log(result.bundle_id) // ev_01j...
console.log(result.summary) // AI-generated summary
console.log(result.manifest_sha256) // tamper-evident hash
console.log(result.key_points_cited[0].citations[0].quote)Python
from docimprint import DocImprintClient
client = DocImprintClient(api_key="dr_live_...")
result = client.extract(
url="https://example.com/contract.pdf",
include=["markdown", "summary"],
)
print(result["bundle_id"]) # ev_01j...
print(result["summary"]) # AI-generated summary
print(result["manifest_sha256"]) # tamper-evident hashREST API (cURL)
curl -X POST https://api.docimprint.com/v1/extract \
-H "Authorization: Bearer dr_live_..." \
-H "Content-Type: application/json" \
-d '{"source": "https://example.com/contract.pdf", "include": ["markdown", "summary"]}'Features
Capability | Method / Mode | Description |
Extract |
| Full evidence bundle — markdown, summary, cited key points, artifacts, manifest SHA-256 |
Invoice parsing |
| Structured |
Document comparison |
| Diff two documents; supply |
Structured extraction |
| Extract typed fields using a caller-defined JSON Schema |
Summarize |
| Prose summary + key points, each with cited paragraph references |
Q&A |
| Single-question answer with inline citations and confidence score |
Claim-check |
| Per-claim verdict: |
Translate |
| Source-cited translation to any language |
Describe |
| AI description of image or PDF page contents |
Collections |
| Semantic search + cross-document RAG Q&A across multiple bundles |
Verify |
| Cryptographic integrity check — manifest hash, EIP-191 signature, artifact hashes. Free, no auth |
Notarize |
| Write manifest SHA-256 to Base L2 via EAS — permanent, immutable attestation |
MCP server | 20 tools · 2 prompts | Native MCP for Claude, Cursor, and any MCP-compatible client |
x402 | REST | Pay per call in USDC — no account required |
Why DocImprint?
🔏 Cryptographic provenance — every bundle is EIP-191 signed at creation; optionally notarized on Base L2 via EAS for an immutable on-chain audit trail
🤖 Agent-native by design — async jobs, webhooks, idempotency keys, and legal hold built into every request, not bolted on
📎 Beyond OCR — citations carry exact quotes, paragraph indices, and confidence scores (
high/medium/low), not just raw text🛠️ 10 CrewAI tools out of the box —
research_tools(),legal_tools(),collection_tools()pre-grouped for common agent workflows⚖️ Compliance-ready —
legal_hold, provenance logging, multi-agent handoff tracking, and on-chain notarization designed for legal and regulated industries💳 Flexible payment — monthly credit plans or pay-per-call USDC via x402, no account required
Agent-native features
DocImprint is designed for autonomous agent workflows, not just synchronous API calls.
// Fire-and-forget async extraction — returns job_id immediately
const job = await client.extract({
source: 'https://example.com/large-report.pdf',
async: true,
webhook: 'https://your-agent.io/callback',
idempotency_key: 'report-2025-q4',
legal_hold: true,
})
const status = await client.getJob(job.job_id)
// { status: 'complete', bundle_id: 'ev_...', progress_pct: 100 }
// Monitor a URL for changes — get notified on diff
await client.extract({
source: 'https://example.com/terms.html',
monitor: { webhook: 'https://your-agent.io/changes', mode: 'diff' },
})Python: provenance & multi-agent handoff tracking
client.log_provenance(bundle_id="ev_...", agent_id="agent-research", action="extracted")
client.handoff(bundle_id="ev_...", from_agent="agent-research", to_agent="agent-legal", note="ready for claim check")CrewAI integration
10 purpose-built tools for CrewAI agents, organized into preset groups.
from docimprint.crewai import DocImprintToolkit
toolkit = DocImprintToolkit(
api_key="dr_live_...",
collection_id="col_...", # required for collection tools
)
toolkit.research_tools() # extract, summarize, qa, check_claims
toolkit.legal_tools() # check_claims, verify, notarize
toolkit.collection_tools() # search, ask, add_to_collection
toolkit.all_tools() # all 10 toolsTool | What it does |
| Full evidence bundle with citations and manifest hash |
| Key points with cited paragraph references |
| Cited answer with confidence score |
| Per-claim verdict: supported / contradicted / not_found |
| Source-cited translation |
| Cryptographic integrity check — manifest, hash, signature |
| On-chain EAS attestation on Base L2 |
| Semantic vector search across document collection |
| Cross-document RAG Q&A with cited sources |
| Add bundle to collection and trigger async indexing |
ProvenanceTracker wraps all tools to automatically log agent actions and bundle handoffs.
from docimprint.crewai import DocImprintToolkit, ProvenanceTracker
tracker = ProvenanceTracker(client=toolkit.client)
trackable = toolkit.trackable_tools()DocImprintKnowledgeSource integrates with CrewAI's knowledge system for retrieval-augmented agents.
MCP server
DocImprint exposes a native MCP server for use with Claude, Cursor, and any MCP-compatible client.
Install via Smithery (one command):
npx @smithery/cli install docimprint --client claudeManual config:
{
"mcpServers": {
"docimprint": {
"type": "streamable-http",
"url": "https://api.docimprint.com/mcp",
"headers": {
"Authorization": "Bearer dr_live_..."
}
}
}
}Transport: streamable-http · Auth: Bearer token · Listed on Smithery · Listed on Glama
MCP tools
URL tools — lean mode, no bundle stored:
Tool | Description |
| Fetch URL → text + metadata |
| Fetch URL → prose summary + key points |
| Fetch URL → cited answer to a specific question |
| Fetch URL → translated content |
Document tools — accepts base64 PDF or image:
Tool | Description |
| OCR plain text from PDF / image |
| OCR tables as Markdown from PDF / image |
| Structured invoice fields: merchant, line items, totals + citations |
| Prose summary + cited key points |
| Per-claim verdict with evidence quotes |
| Extract typed fields using a caller-defined schema |
Bundle & collections:
Tool | Description |
| Cryptographic integrity check |
| Bundle metadata and notarization status |
| On-chain EAS attestation on Base L2 |
| Create a named document collection |
| List your collections |
| Add bundle + trigger async indexing |
| Semantic vector search |
| Cross-document RAG Q&A |
| Poll async job (extract, indexing, batch) |
| Check credit balance and plan status |
Guided prompts: claim_check_workflow · invoice_intake
Resource: bundle://{bundle_id} — read bundle metadata directly
API reference
Full typed reference: docimprint.com/docs · OpenAPI
Category | Methods |
Core |
|
Focused |
|
Extract modes |
|
Collections |
|
Jobs |
|
Error handling
import { DocImprintClient, DocImprintError } from 'docimprint'
try {
const result = await client.extract({ source: 'https://example.com/doc.pdf' })
} catch (err) {
if (err instanceof DocImprintError) {
console.error(err.message) // human-readable error
console.error(err.status) // HTTP status code
console.error(err.requestId) // x-request-id for support
}
}Authentication
API key
Monthly credits via Stripe — sign up at docimprint.com, free tier available:
const client = new DocImprintClient({ apiKey: 'dr_live_...' })x402 USDC payments
DocImprint supports the x402 open standard — pay per call in USDC on Base, straight from any EVM wallet. No account. No sign-up. No API key. Your agent can call the API autonomously without any human-managed credentials.
The API is x402-native: it returns a standard 402 Payment Required response with on-chain payment details, your client settles in USDC, and the request completes automatically. From $0.01 / call.
With @x402/fetch (automatic payment handling):
import { wrapFetchWithPayment } from '@x402/fetch'
import { createWalletClient, http } from 'viem'
import { privateKeyToAccount } from 'viem/accounts'
import { base } from 'viem/chains'
const account = privateKeyToAccount('0x...')
const wallet = createWalletClient({ account, chain: base, transport: http() })
const fetchWithPayment = wrapFetchWithPayment(fetch, wallet)
const res = await fetchWithPayment('https://api.docimprint.com/v1/summarize', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ source: 'https://example.com/report.pdf' }),
})
const data = await res.json()Identity on owner routes: For GET/DELETE operations that require ownership (e.g. fetching your bundle), pass the payment-signature header from a prior payment response so the API can verify your wallet address without a separate login.
TypeScript
All request and response types are exported:
import type {
ExtractRequest,
ExtractResponse,
Citation,
CitedField,
InvoiceResult,
Job,
Collection,
SearchResult,
} from 'docimprint'Pricing
Free tier available — no credit card required. Monthly credit plans via Stripe, or pay per call in USDC via x402. See pricing →
Community
Questions, integrations, and announcements: GitHub Discussions.
Links
This server cannot be installed
Maintenance
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/sawftware-apps/docimprint-sdk'
If you have feedback or need assistance with the MCP directory API, please join our Discord server