kiro-validate-impl

Cross-task integration validation verifies that completed tasks respect design boundaries and data contracts by mapping task outputs to consuming tasks, running the full test suite, and auditing boundary violations against the design. It produces a GO or NO-GO remediation report to confirm feature completion.

Requirements coverage and runtime liveness checks map requirements.md sections to completed tasks and discover canonical test and smoke commands to run repository-level tests. This maps feature requirements to implemented tasks and confirms the system boots and passes tests before release.

Post-implementation boundary scans grep feature boundaries for TODO, TBD, and HACK markers alongside hardcoded secrets. The results are classified by ownership and included in a remediation report to ensure code hygiene before final integration approval.

Yes, cross-task consistency validation is designed for features stored under .kiro/specs. It checks completed tasks within this directory structure for data-shape mismatches, boundary violations, and requirements coverage during the post-implementation phase.

Per-task tests only verify individual task outputs, while full-suite integration checks discover cross-task contract mismatches and boundary violations. Features can pass isolated per-task checks but fail full-suite validation due to data-shape mismatches between consuming tasks or design boundary infractions.

A cross-task integration validator checks runtime liveness, cross-task contract consistency, requirements coverage, and code hygiene during post-implementation. It discovers test commands, maps requirements to tasks, scans for TODOs and secrets, and generates ownership-classified remediation guidance.