Documents
What it is
Section titled “What it is”Documents are files uploaded to a program for AI analysis. They serve as the raw input for the Discovery pipeline — PDFs, spreadsheets, Word documents, and other formats are stored, analyzed by Claude, and used to extract structured findings.
Why it matters
Section titled “Why it matters”Project knowledge lives in documents. By centralizing uploads and running AI extraction, Foundry converts static files into structured, actionable requirements without manual transcription.
Key concepts
Section titled “Key concepts”- Upload: Drag-and-drop or click-to-upload via the Discovery document zone
- Storage: Files are stored in Convex file storage with metadata in the
documentstable - Analysis: Each document can be analyzed by AI to extract findings (requirements, risks, decisions)
- Re-analysis: Previously analyzed documents can be re-processed with
ReAnalyzeDialog - Status tracking: Documents move through
uploaded,analyzing,analyzed,errorstates - Program scope: Documents belong to a program and respect org-level access control
How to use it
Section titled “How to use it”- Navigate to [Program] > Discovery > Documents.
- Drag files into the upload zone or click to browse.
- Wait for upload completion — the document appears in the list.
- AI analysis runs automatically; findings appear in the Findings tab.
- Re-analyze documents if new extraction capabilities are available.
Data model
Section titled “Data model”This feature uses the following tables:
documents— File metadata, storage reference, program scope, and analysis statusdocumentAnalyses— AI analysis records per document with extracted contentdiscoveryFindings— Findings extracted from document analysis