kiro-review

Review task implementations against specifications and return APPROVED or REJECTED verdicts.

3.6k|274|Updated Jul 17, 2025
One-click install
npx skills add https://github.com/gotalab/cc-sdd --skill kiro-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kiro-review
Source: https://github.com/gotalab/cc-sdd/tree/main/tools/cc-sdd/templates/agents/claude-code-skills/skills/kiro-review
Command: npx skills add https://github.com/gotalab/cc-sdd --skill kiro-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ensures task implementations are real, bounded, and aligned with approved requirements and design while providing mechanical verification evidence to avoid hidden scope creep or silent deviations from the spec.

Core Features & Use Cases

  • Mechanical verification: inspect diffs, run the canonical test suite, run lint/static checks where available, and grep for placeholders or secrets.
  • Boundary and design alignment: compare changed files to task Boundary: constraints and referenced design and requirement sections to detect spillover or hidden dependencies.
  • Use Cases: run after an implementer marks READY_FOR_REVIEW, after remediation for rejected work, or before accepting a task into higher-level validation.

Quick Start

Run a kiro-review for task T-123 by providing the task id, exact task text, spec file paths and section refs, the implementer status report, boundary scope, and validation commands so the reviewer can inspect diffs, run tests, check for TODOs/secrets, and emit an APPROVED or REJECTED verdict.

Frequently Asked Questions about kiro-review

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

FAQPage Schema
How do I verify code changes against specifications before marking a task complete?

Spec compliance verification inspects code diffs, runs canonical test suites, and checks for placeholders to ensure task implementations align with approved requirements and boundaries. This review process detects hidden scope creep and silent deviations before acceptance.

What is adversarial code review and how does it prevent scope creep?

Adversarial code review mechanically verifies implementations against approved specifications and task boundaries by running static checks and searching for hardcoded secrets. It prevents scope creep by auditing changed files to detect spillover or hidden dependencies not defined in the spec.

How do I check for hardcoded secrets and placeholder markers in a codebase?

Checking for hardcoded secrets and placeholders involves greping changed files for TODO markers and sensitive strings during the review phase. This mechanical verification ensures implementations contain real logic rather than unfinished fragments before approval.

When should I run static analysis and boundary audits in a spec-driven workflow?

Static analysis and boundary audits should run after an implementer reports READY_FOR_REVIEW, after remediation, or before marking a task complete. This timing ensures all code changes are mechanically verified against design constraints prior to higher-level validation.

Can I use automated review to generate APPROVED or REJECTED verdicts for task implementations?

Automated review generates structured APPROVED or REJECTED verdicts by comparing changed files to task boundaries, running validation commands, and auditing design alignment. The verdict includes specific findings and remediation steps for rejected work.

What's the best way to audit task boundary adherence in spec-driven development?

Auditing task boundary adherence requires comparing changed files to task Boundary constraints and referenced design sections to detect spillover. This review confirms implementations stay within approved scope and avoids hidden dependencies before final acceptance.