Documentation

DRIFT Overview

Last updated: 27 March 2026

DRIFT is a Jira app for requirements traceability and backlog review. It helps teams preserve a source requirement baseline, review Jira work against those requirements, and make human decisions before stories move into delivery.

Core workflow

• Ingest source requirements from pasted text, documents, or Confluence pages
• Review parsed requirements and coverage at the project level
• Draft Jira stories directly from uncovered requirements
• Link each Jira issue to one primary requirement
• Run validation and record a review decision

What DRIFT does today

• Requirement ingestion from text, files, and Confluence
• Deterministic drift validation for numeric, timing, and coverage issues
• Optional AI semantic review
• AI-assisted requirement extraction for messy source text
• Requirement coverage tracking based on explicitly linked Jira issues
• Draft-story generation from a requirement with editable acceptance criteria

Licensing

DRIFT is maintained as two Forge app variants from one shared codebase:

• Beta app: private testing and design-partner usage
• Marketplace app: licensed Paid via Atlassian version for customer installs

Recommended use case

DRIFT is a good fit for teams that generate or refine Jira backlog items from client requirements, internal specifications, Confluence pages, or AI-assisted story drafting workflows and want a quality gate before engineering starts.

Important notes

• DRIFT outputs support human review and are not automatic delivery approval
• AI-assisted features are optional and require provider configuration
• Confluence import depends on access to the target page on the same Atlassian site
• Coverage is based on explicit links, not fuzzy matches