How ensure consistency of new specs and plans with the previous ones? #974

There are no clear instructions about how to develop with spec-kit when multiple features/specs are created over the time. It seems the method only works under the assumption that every spec is independent and iterations only happen in the context of a single spec. In big projects or incremental features, each new spec and plan does not consider the previous ones and lot of decisions are lost, breaking consistency across the project.

I've attempted to address this problem by enhancing the /analyze command to validate
cross-spec consistency.

Sample:

   - **Cross-spec inventory**: Build comprehensive index from all previous specs/plans:
     - Component names and their responsibilities
     - Data entities/models with their fields and types
     - API patterns and conventions
     - Technology choices and architectural patterns
     - Shared interfaces, integration points, and dependencies between components
     - Naming conventions and terminology used consistently across specs

Before I invest more time in this approach, I wanted to ask:

1. Is this a known limitation that's on the roadmap?
2. Would a cross-spec validation feature align with spec-kit's design philosophy?
3. Are there alternative approaches you'd recommend?