kiro-validate-impl

Validates feature-level integration across completed tasks with test suites, coverage, and design alignment checks.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/scapia-oss/compass --skill kiro-validate-impl-scapia-oss
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kiro-validate-impl
Source: https://github.com/scapia-oss/compass/tree/main/tools/cc-sdd/templates/agents/claude-code-skills/skills/kiro-validate-impl
Command: npx skills add https://github.com/scapia-oss/compass --skill kiro-validate-impl-scapia-oss

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? After individual tasks are implemented and reviewed in isolation, cross-task problems like integration seams, duplicated logic, requirement gaps, and design drift remain invisible. This Skill performs a whole-feature validation pass that catches issues only visible when all completed tasks are examined together. ## Core Features & Use Cases - Mechanical Gates: Runs the full test suite, greps for residual TODOs and hardcoded secrets, and executes a runtime smoke-boot check, returning GO, NO-GO, or MANUAL_VERIFY_REQUIRED. - Cross-Task Judgment Checks: Verifies interface contracts, shared state consistency, cross-unit code duplication, requirements coverage percentage, and end-to-end design alignment including spec-drift scans. - Shared Component Regression: Optionally dispatches a regression-verifier subagent when shared code with external callers was modified, with a --regression flag to force the check. - Use Case: After finishing all tasks of a user-auth feature, run the validation to confirm the full test suite passes, every requirement section maps to a completed task, and the implementation still matches the approved design before handoff. ## Quick Start Ask the agent to validate the completed feature by running kiro-validate-impl with the feature name, for example to validate user-auth end to end.

Frequently Asked Questions about kiro-validate-impl

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I validate a feature after all tasks are implemented?

Run the validation command with the feature name, for example /kiro-validate-impl user-auth. It detects completed tasks, runs the full test suite and smoke checks, audits cross-task integration and design alignment, then writes a GO or NO-GO report to impl-validation.md.

What is the difference between task-level review and feature-level validation?

Task-level review during implementation checks individual acceptance criteria and per-file reality. Feature-level validation only examines cross-task concerns: integration seams, requirements coverage across all tasks, design drift, duplication between units, and shared-component regressions.

When should I use the --regression flag?

Use --regression to force the shared-component regression check even when no shared code change is auto-detected. Without the flag, the check runs only when the feature modified shared code with callers outside its boundary.

What does MANUAL_VERIFY_REQUIRED mean in the validation report?

It means a mandatory check could not be completed, such as an unknown test command, unavailable runtime environment, or an environmental failure like expired credentials. The feature must not be treated as complete until the missing validation step is performed.

Does this validation replace running the test suite in CI?

No. It runs the repository's canonical test and smoke commands as one signal among many, and adds judgment checks for requirements coverage, design alignment, and boundary integrity that CI test runs do not evaluate.