Understand every document.
Trace every insight.
Evidence-grounded document analysis you can verify. Extract deadlines, obligations, financial values, and potential anomalies with source pages and locations.
PDF Ingestion & Structured Analysis
Uploading PDF
PDF selected and sent to the FastAPI analysis endpoint for processing.
Extracting Content
Document text is extracted while preserving page references so results can be traced back to their source.
Analyzing Document
The document is analyzed for document type, deadlines, financial values, obligations, and potential review items.
Preparing Dashboard
Results are returned as structured JSON with extracted fields, source pages, evidence, and rule-based confidence estimates.
Evidence: From Result to Source Page
Each extracted result can be traced back to the source page and text used for the analysis.
Payment Deadline
Loading analysis…
Connecting to the DocTrace backend.
Loading analysis…
Connecting to the DocTrace backend—
—Selected Risk & Anomaly Checks
Rule-based checks look for selected missing fields in supported contract and invoice types, conflicting detected dates or contract/payment amounts, and arithmetic mismatches when a subtotal, tax, and total are all detected. Review each result in the original PDF.
Examples below are from the bundled demo PDF and are for demonstration only.
Amount Checks
Extract financial values and compare subtotal, tax, and total when all three are detected.
Payment Deadlines
Extract date patterns and payment periods such as within 30 days with a source location for review.
Key Obligations
Find sentences containing common obligation cues such as shall, must, or will so reviewers can check the full clause.
Consistency Checks
Flag multiple detected values for selected date and amount fields, or a subtotal-plus-tax mismatch when all figures are found.
DocTrace is an automated review aid, not legal advice. Extraction and rule-based checks can miss or misread content. Confidence scores are heuristic estimates, not calibrated probabilities. Verify results in the original PDF and consult a qualified lawyer for legal decisions.
Ask about the analyzed document.
DocTrace searches the uploaded PDF text for matching passages and returns a source page when it finds relevant text. It does not interpret legal effect.
Ask DocTrace Assistant
Sample conversation • demo document
Within 30 days of receiving the invoice after milestone acceptance. The clause appears on page 7, Section 4.2.
The sample flags conflicting effective dates and contract values on page 8, plus an invoice total that does not match its subtotal and tax on page 11.
Built on HTML, Bootstrap, JavaScript & FastAPI
style.css (DocTrace visual system)app.js (Upload, API connection & dashboard data)index.html (Landing & product overview)upload.html (PDF upload)dashboard.html (Analysis results)The frontend sends the selected PDF to FastAPI for document analysis and receives structured JSON for the dashboard.
DocTrace Analysis Flow
1. Upload the PDF
The user selects a document from the landing or upload page.
2. Analyze the document
FastAPI extracts document content and sends it through the analysis pipeline.
3. Review structured results
The dashboard presents rule-based extracted fields, potential review items, source pages, and heuristic confidence estimates.
Turn every PDF into
structured, traceable results.
Upload a document to extract its type, deadlines, amounts, obligations, risks, source pages, and confidence through the DocTrace analysis workflow.
