kiro-verify-completion

Verify TASK, FIX, TEST_OR_BUILD, and FEATURE_GO claims against current code state.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill ensures claims about work items are verified against fresh evidence before marking completion, preventing misrepresentation and premature progress.

Core Features & Use Cases

  • Evidence-driven verification: validates TASK, FIX, TEST_OR_BUILD, and FEATURE_GO claims with current code state.
  • Structured outcomes: returns VERIFIED, NOT_VERIFIED, or MANUAL_VERIFY_REQUIRED with supporting evidence.
  • Guardrails for autonomous execution: reduces risk of incorrect GO decisions and helps maintain quality.

Quick Start

Ask the verifier to check a specific claim against the current code state and report a clear verdict.

Frequently Asked Questions about kiro-verify-completion

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

FAQPage Schema
How do I verify task completion claims against fresh code state?

Task completion verification validates claims by executing canonical validation commands against current code state, returning structured verdicts like VERIFIED or NOT_VERIFIED with supporting evidence. This prevents premature progress marking.

What is evidence-driven verification for autonomous AI agents?

Evidence-driven verification for autonomous AI agents validates work item claims like TASK, FIX, or FEATURE_GO by inspecting current code state. It provides fresh evidence and guardrails to reduce risks of incorrect autonomous GO decisions.

How do I get a structured verification verdict for a code fix?

To get a structured verification verdict for a code fix, execute a canonical validation command for the FIX claim type. The system returns VERIFIED, NOT_VERIFIED, or MANUAL_VERIFY_REQUIRED with supporting details and exit codes.

Does evidence-driven verification work with test or build claims?

Yes, evidence-driven verification supports TEST_OR_BUILD claims by executing canonical validation commands and inspecting current code state. It returns structured results with exit codes and outputs confirming the test or build status.

When do I need manual verification for task claims?

Manual verification for task claims is needed when automated evidence is insufficient to return a clear VERIFIED or NOT_VERIFIED status. The system returns MANUAL_VERIFY_REQUIRED, prompting human review of supporting details and defined boundaries.