Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
AWS’s contract-intelligence example combines two paths: agents extract and independently check defined fields from contract PDFs, while the original documents remain available for focused questions. Verified fields go into Aurora PostgreSQL for portfolio analytics through Amazon Quick; document retrieval serves questions about a particular contract. The separation matters: a system that finds a relevant passage is not automatically equipped to calculate a reliable total across an entire contract collection.
Why contract intelligence needs both structured data and document retrieval
Consider three questions a contract team might ask:
- “Which vendor are we spending the most with?”
- “Which contracts are about to expire?”
- “What are the payment terms in the AnyCompany contract?”
The first two require comparable values across a portfolio. The third asks about the contents of one source document. Those are different retrieval problems, so the architecture uses complementary approaches.
Retrieval-augmented generation (RAG) can retrieve relevant passages for a focused lookup. But selecting a limited number of semantically relevant chunks does not guarantee that every contract needed for a portfolio total or comparison is represented. Extracting fields into structured records enables database queries across the collection; retaining the PDFs supports document-grounded questions. Neither path replaces the other.
#1 Best Overall
- PORTABLE SCANNER FOR USE ON-THE-GO — The fastest and lightest mobile single-sheet-fed compact document scanner in its class¹
- QUICK DOCUMENT SCANNING ― This Epson ultra-fast scanner scans a single page as quickly as 5.5 seconds²; Windows and Mac compatible
- VERSATILE PAPER HANDLING ― Portable scanner scans documents up to 8.5 x 72 in; Also easily digitizes receipts and ID cards to make accounting, bookkeeping, and organizing simpler
- INTUITIVE, HIGH-SPEED SOFTWARE — Epson ScanSmart Software³ is a smart tool allowing you to easily scan, review, and save; Stay organized easily with the help of this Epson scanner
- EASY SETUP — USB-powered connect to your computer for quick and simple scanning; No batteries or external power supply required to operate portable document scanner; Standard Connectivity: USB 2.0
AWS frames the problem in its September 29, 2026 Machine Learning Blog article by Konala McGrath, Hugo Tse, Alberto Alonso, and Nitish Chaudhari: “A better prompt won’t fix this. A different architecture will.” The practical point is not that prompts or RAG are useless, but that a portfolio-wide computation needs an architecture that makes the relevant records available for computation.
How the AWS extraction-and-verification pipeline works
The September 2026 example is an AWS reference architecture, not a general performance benchmark. Its core path preserves the contract PDFs, extracts a defined set of fields, checks those fields with another model, and stores verified data for analytics.
Rank #2
- Fast and Efficient: Scans both sides of a document at the same time, in color, at up to 45 pages per minute, with a 60 sheet automatic feeder, and one touch operation. Innovative Feeding System.
- Reliably Handles Many Different Document Types: Receipts, business cards, reports, contracts, long documents, thick or thin documents, and more. Monochrome LCD Display.
- Designed exclusively for the included Canon CaptureOnTouch software;TWAIN and ISIS drivers are not supported.
- Easy Setup: Simply connect to your computer using the supplied USB-C cable.
- Bundled Software: Includes easy-to-use Canon CaptureOnTouch scanning software.
- Keep the original PDFs in Amazon S3. The source files remain available for checking extracted values and answering questions about individual contracts.
- Extract a defined schema. An extraction agent reads each PDF and returns eight fields, each with a confidence score. The example uses a Claude Sonnet-series model and native PDF reading. It is a specific implementation choice, not a requirement for every deployment; model options and regional availability change, so confirm what is available in the intended AWS Region.
- Ask a separate agent to check the fields. A verification agent reads the same contract and checks the extracted values. AWS’s rationale is that a different model can offer an independent perspective. If the models disagree, treat that as a signal to review the field—not proof that the verifier is correct.
- Use Textract for the narrow signature dispute. In this example, Textract performs visual signature detection only when the two models disagree about the
is_signedfield. AWS notes that a model may mistake an empty signature line for an actual signature. This targeted check concerns whether a signature appears visually; it does not determine what a clause means or resolve every question of contract interpretation. - Save verified fields in Aurora PostgreSQL. Structured records make it possible to query and aggregate extracted values across the collection. Keep the original PDFs available for source-grounded lookups.
- Serve portfolio and document questions through the appropriate path. The example presents analytics and natural-language querying through Amazon Quick. Use structured records for totals and comparisons, and refer to the relevant source document for a question about its wording.
Two different ways to divide work among agents
“Multi-agent” does not describe just one workflow. The September field-verification example and a separate AWS contract-management guide published January 27, 2026 divide work in different ways.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →| Design | How the work is divided | What it is for |
|---|---|---|
| Independent field verification | An extraction agent produces defined fields; another agent checks them. Textract is invoked for the specific visual signature dispute described in the example. | Checking extracted values before they are stored as verified data. |
| Business-function collaboration | A collaboration agent coordinates legal, risk, and compliance agents, then consolidates their findings. | Bringing different kinds of contract-management analysis into one workflow. |
In the January guide, the legal agent extracts parties, terms, and obligations; the risk agent assesses financial and operational exposure; and the compliance agent evaluates regulatory requirements. The guide connects Quick Suite workflows and data access with AgentCore agents, S3 documents, and Redshift structured data. This function-based arrangement is not the same as asking a second model to independently verify extracted fields.
Rank #3
- Scanner type: Document
- Connectivity technology: USB
- With Auto Scan Mode, the scanner automatically detects what you're scanning
- Digitize documents and images
What AgentCore provides—and what it does not
AWS describes AgentCore as modular infrastructure for building agents with multiple frameworks and foundation models. Its components support parts of an agent workflow:
- Runtime: hosting and scaling agent workloads.
- Identity: agent identity and access.
- Gateway: making APIs and tools available to agents.
- Code Interpreter: sandboxed code execution.
- Observability: tracing, auditing intermediate outputs, and debugging workflow performance.
- Harness, Memory, and Browser: additional capabilities listed in the official AgentCore documentation.
These are platform functions, not guarantees of correct extraction, authorized access to every document, or legal and regulatory compliance. Those outcomes depend on the application’s permissions, data handling, evaluation, and oversight.
Rank #4
- OUR MOST ADVANCED SCANSNAP. Large touchscreen, fast 45ppm double-sided scanning, 100-sheet document feeder, Wi-Fi and USB connectivity, automatic optimizations, and support for cloud services. Upgraded replacement for the discontinued iX1600
- CUSTOMIZABLE. SHARABLE. Select personalized profiles from the touchscreen. Send to PC, Mac, mobile devices, and clouds. QUICK MENU lets you quickly scan-drag-drop to your favorite computer apps
- STABLE WIRELESS OR USB CONNECTION. Built-in Wi-Fi 6 for the fastest and most secure scanning. Connect to smart devices or cloud services without a computer. USB-C connection also available
- PHOTO AND DOCUMENT ORGANIZATION MADE EFFORTLESS. Easily manage, edit, and use scanned data from documents, receipts, photos, and business cards. Automatically optimize, name, and sort files
- AVOIDS PAPER JAMS AND DAMAGE. Features a brake roller system to feed paper smoothly, a multi-feed sensor that detects pages stuck together, and skew detection to prevent paper damage and data loss
AWS’s sample repository is another design reference. It describes agents for contract Q&A, administration, analytics, contract creation, compliance, and renewals, plus role-based access. Its listed stack includes AgentCore Runtime, Bedrock, Knowledge Bases, OpenSearch Serverless, S3, Cognito, and CloudFront. Treat sample code as an implementation example, not as a tested guarantee for another organization’s data, controls, or production workload.
How to evaluate the approach for your contracts
AWS reports hand-labeling eight fields across 20 contracts—a total of 160 field values—to compare extractor and verifier combinations. That is a small, directional evaluation, not evidence of general accuracy or legal correctness. AWS also notes that results can vary with contract format and field complexity and recommends local evaluation against the builder’s own benchmarks and success criteria.
Best Value
- FITS SMALL SPACES AND STAYS OUT OF THE WAY. Innovative space-saving design to free up desk space, even when it's being used
- SCAN DOCUMENTS, PHOTOS, CARDS, AND MORE. Handles most document types, including thick items and plastic cards. Exclusive QUICK MENU lets you quickly scan-drag-drop to your favorite computer apps
- GREAT IMAGES EVERY TIME, NO EXPERIENCE REQUIRED. A single touch starts fast, up to 30ppm duplex scanning with automatic de-skew, color optimization, and blank page removal for outstanding results without driver setup
- SCAN WHERE YOU WANT, WHEN YOU WANT. Connect with USB or Wi-Fi. Send to Mac, PC, mobile devices, and cloud services. Scan to Chromebook using the mobile app. Can be used without a computer
- PHOTO AND DOCUMENT ORGANIZATION MADE EFFORTLESS. ScanSnap Home all-in-one software brings together all your favorite functions. Easily manage, edit, and use scanned data from documents, receipts, business cards, photos, and more
For a useful evaluation, build a representative set of your own agreements and record the correct value for each field. Include the document formats and field cases that matter in practice, then measure extraction and verification against that ground truth. Decide in advance which disagreements or low-confidence cases need human review. A second model creates a check and a review signal; it does not remove the need to establish what counts as correct.
- Test the fields you actually intend to query, including difficult or ambiguous cases.
- Track field-level results rather than treating a document as simply “correct” or “incorrect.”
- Set a clear escalation route for disagreements and uncertain values.
- Check that a reviewer can trace a stored field back to its source document.
- Retest when model choices, document formats, or the workflow change.
Decisions to make before deployment
The AWS examples outline architectural patterns; they do not establish that a particular deployment is suitable for every contract operation. Work through these choices against your documents and controls:
Quick Recap
- Portfolio analytics or single-document lookup? Use structured records for cross-contract totals and comparisons; retain source-document retrieval for focused questions.
- Extraction alone or extraction plus verification? Verification adds an independent check, but its value should be measured against labeled contracts and a defined review policy.
- Model judgment or deterministic document analysis? The example uses Textract for the narrow visual question of whether a signature appears after a model disagreement. Do not extend that use into a claim about legal meaning.
- One coordinated workflow or specialist agents? A coordinator can route work to legal, risk, and compliance functions, while separate verification agents check field values. Choose based on the work to be divided.
- What access and retention rules apply? Define who may access source PDFs and extracted records, how long they are retained, and how human review is handled. Features or sample role controls do not make every deployment compliant by default.
- What will the workflow cost to operate? AWS services can incur infrastructure costs; estimate them for the services and workload you plan to deploy, and check current AWS pricing. The January guide includes cleanup steps because deployed resources have associated costs.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
