Skip to content

Requirements

Requirements are the atomic units of scope in Foundry. Each requirement describes a capability the delivery engagement must produce. Requirements are linked to workstreams, assigned priorities, and annotated with fit-gap classifications that determine how much custom development is needed.

Requirements drive every downstream activity — task creation, skill execution, risk assessment, and sprint planning all reference requirements. Keeping them structured and up-to-date prevents scope drift and gives AI agents the context they need for accurate analysis.

  • Reference ID: A human-readable identifier (e.g., FOUND-01, DISC-03) for cross-referencing
  • Priority: critical, high, medium, low
  • Fit-gap classification: native, config, custom_dev, third_party, not_feasible — determines implementation strategy
  • Status: Tracks whether a requirement has been reviewed, approved, or deferred
  • Workstream assignment: Links each requirement to a workstream for organizational grouping
  • Detail panel: Slide-out panel showing full requirement details, linked tasks, and edit history
  • CSV export: Export requirements to CSV for external reporting
  1. Navigate to [Program] > Requirements.
  2. Browse the requirements table with sortable columns for priority, status, and workstream.
  3. Click any row to open the detail panel with full content and linked data.
  4. Use the row action menu to edit, delete, or change status.
  5. Click Export CSV to download the current view.
  6. Create new requirements manually, or let the Discovery feature extract them from documents.

This feature uses the following tables:

  • requirements — Core requirement record with refId, title, description, priority, fitGap, status, workstreamId
  • discoveryFindings — AI-extracted findings that can be promoted to requirements