Source Management

Organize and reuse documents as data sources across any form filling session

Overview

When you add a document to the source library, it is available for any form filling session in your workspace without re-uploading. Upload a physician CV, a client intake packet, or a prior completed form once, and any future session that needs it can pull from the library directly.

This is different from how profiles work. Profiles store structured data - names, addresses, dates - that maps directly to form fields. Source library documents are unstructured files that the AI reads and interprets to extract values. In practice, most workflows use both: the profile handles the standard identity fields while source documents carry case-specific detail like prior applications, employment records, or supporting certificates.

The library supports tagging, folder organization, version history, and usage tracking across sessions. Workspace-level sharing gives teams a single, maintained document set rather than individual uploads duplicated across accounts.

Key Capabilities

• Source Library: Centralized repository for frequently-used documents, scoped to workspace
• Multiple Format Support: PDF, Word, Excel, images, text files, and plain text
• Source Organization: Tag and folder organization for easy discovery
• Version History: Track changes when updating source documents; prior versions are preserved
• Usage Tracking: See which sources are used in which sessions
• Quick Access: Recently-used sources appear first for easy selection
• Bulk Upload: Upload multiple source documents at once for batch processing workflows
• Source Sharing: Share sources with workspace members for team collaboration
• OCR for Images: Google Cloud Vision API extracts text from scanned documents and photographed IDs
• Source Metadata: Add descriptions, tags, and notes to each source

How It Works

1. Add Sources to Library:

   • Upload documents from your computer
   • Paste text directly
   • Capture via email forwarding
   • Add programmatically via RESTful API: POST /api/sources

2. Processing on Upload:

   • Text extracted from PDF (PyMuPDF), Word, Excel, and plain text
   • Images and scanned PDFs processed via Google Cloud Vision API (OCR), with PIL/Pillow and OpenCV 4.9.0 for pre-processing (rotation correction, contrast normalization)
   • Vector embeddings created per source; source text mapped to form page numbers for page-scoped retrieval during sessions
   • Text content encrypted with workspace-scoped keys before storage

3. Organize Sources:

   • Create folders: "Client Documents", "Reference Materials", "Templates"
   • Tag sources: "resume", "financial-info", "medical-history"
   • Add descriptions for team members

4. Use in Form Sessions:

   • Start a form filling session
   • Select "Add Source" - "From Library"
   • Sources are available to the session before autofill begins
   • AI extracts data automatically from each source for the relevant fields

   Example: Teams using the n8n integration push source documents from Google Sheets, CRMs, or email inboxes into Instafill.ai via the API, so sessions start with sources already attached and no manual upload step is needed.

5. Update Sources: When information changes:

   • Upload an updated version
   • The system creates a new version while preserving previous versions
   • Future sessions use the latest version automatically
   • In-progress sessions continue with the version they started with, so a mid-session update does not affect an active fill

6. Share with Team:

   • Mark sources as "Shared" for workspace visibility
   • Team members reference common sources (company info, standard templates)
   • No duplicate uploads needed

Use Cases

Source management saves the most time in workflows where the same documents appear across many form submissions.

Medical credentialing - Hawkeye Physicians maintains a source library per radiologist containing CME certificates, malpractice coverage letters, and board certification records. When a new hospital contract requires a credentialing packet, the relevant sources are already organized and ready - no hunting across email or shared drives. Time per packet dropped from 3-4 hours to under 30 minutes.

Insurance authorization - Headstart Health processes ABA treatment authorization forms for 30+ payers. Each form pulls data from 3-5 documents: treatment plans, diagnostic reports, insurance cards, provider rosters. Keeping those in a shared source library means any team member can start a session without tracking down files. Time per form dropped from 20-30 minutes to around 2 minutes.

Immigration law - For firms managing multiple USCIS forms per client, source documents like passports, birth certificates, employment letters, and prior applications are uploaded once and reused across all filings in the case. Hong LLC achieved a 75-80% time reduction across I-485, I-765, I-130, and G-1145 filings this way.

HR and onboarding - Employee documents (I-9 supporting IDs, certifications, prior W-2s) are uploaded once per employee and reused across any onboarding or compliance form without the HR team re-uploading per form. When a new certification is earned, only one source needs updating for all future sessions to reflect it.

Benefits

• Time savings: Upload documents once, use across any number of sessions without re-uploading
• Consistency: The same source used across multiple forms ensures data extracted from it stays uniform
• Version control: Update a source without losing previous versions
• Team collaboration: Share sources across a workspace so every team member draws from the same maintained document set
• Reduced errors: Referencing a single source eliminates transcription mistakes that occur when the same data is manually re-entered per form
• Easy updates: Update a source once and all future sessions use the new version automatically
• Better organization: Tag and folder structure keeps documents findable rather than buried in email threads or local folders

Security & Privacy

• Access Control: All service-layer queries include a workspaceId filter - source data is isolated per workspace and protected via the shared JWT authentication middleware running in both the .NET and Python service layers
• Encryption: Text source content encrypted with workspace-scoped keys stored in Azure Key Vault; binary files stored in Azure Blob Storage
• Scope Restriction: Encryption includes scope metadata that prevents decryption outside the originating workspace, even for internal service calls
• Retention Policies: Configure automatic deletion after a specified period via Data Retention & Cleanup Management
• Audit Trail: Every access and modification to sources is logged - see Request Logging & Audit Trail for details
• No AI Training: Source documents are never used to train AI models; they are processed only for your specific form filling sessions

Common Questions