aif-improve

Plan refinement analyzes your existing plan against the codebase to identify gaps, misconfigurations, and missing tasks. It applies project-wide context rules and performs deep codebase analysis to ensure tasks are correctly scoped, ordered, and aligned with your architecture.

Plan dependencies become misaligned when the original plan lacks deep codebase analysis. Re-analyzing the plan against the current codebase enforces plan integrity by updating dependencies, preserving completed work, and ensuring tasks are correctly ordered for safe, incremental improvements.

The best way to refine a development plan while preserving completed work is to re-analyze it against the codebase to identify gaps. This process enforces plan integrity by updating dependencies and generating a concise refinement report that guides implementation.

To ensure safe, incremental improvements, you can validate task ordering through dependency and task-order validation during plan refinement. This process checks the existing plan against the codebase to confirm tasks are correctly scoped and ordered according to project conventions.

Yes, AI-assisted planning detects missing tasks by performing deep codebase analysis to compare the existing plan against the actual implementation. This identifies gaps and misconfigurations, generating an actionable report that guides your next implementation steps.

You should avoid automated plan refinement when you lack an existing plan to evaluate, as this process specifically re-analyzes and polishes current plans. It requires an initial plan and project context to correctly scope tasks and update dependencies accurately.