ScopeLens
An evidence-first requirement change workbench that preserves source context, versions business rules, maps affected work, and keeps impact decisions human-reviewed.
- Team
- Independent project
- Period
- 2026
- Status
- Shipped
- Platform
- Web
- Languages
- TypeScript · CSS
The problem
A small requirement change can ripple across an API rule, a boundary test and a screen, but the original wording and the evidence behind those connections are easily lost. Teams need a reviewable way to understand that impact before implementation begins.
How it was built
- 01
Preserved the original brief and exact source quotations alongside reviewed requirements, so every saved rule retains its evidence context.
- 02
Maintained immutable requirement versions with change reasons and optimistic concurrency checks, preventing stale edits from silently replacing newer decisions.
- 03
Connected requirements to APIs, tests, screens and related requirements through explicit evidence links, then captured linked-impact snapshots for review.
- 04
Kept every impact decision human-reviewed: suggestions can be accepted, dismissed or reopened, and decisions remain visible in the audit history.
- 05
Bound optional AI suggestions to supplied artifacts, schema validation and exact source quotation; linked analysis remains usable without an AI provider.
- 06
Built the React and TypeScript interface over a NestJS service with PostgreSQL, using embedded PGlite locally and resettable fictional demonstration data on the public deployment.
What shipped
A working requirement-change flow from source brief and versioned rule to evidence-linked impact review.
Human decisions and version staleness remain explicit; accepting an impact does not modify code, tests or external systems.
The public deployment demonstrates the interface with fictional, resettable data rather than presenting itself as a persistent customer platform.
In one paragraph
ScopeLens helps developers understand what happens when a business requirement changes. It preserves exact source evidence, maintains immutable requirement versions, connects APIs, tests, screens, and related work, and presents possible impacts for explicit human review. Optional AI suggestions remain bounded by schema validation and exact evidence quotation.