sdd-verify

Validate implementation against specs, design, and tasks through real test execution.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/Balthael/ciberbal-ai --skill sdd-verify-balthael
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-verify
Source: https://github.com/Balthael/ciberbal-ai/tree/main/internal/assets/skills/sdd-verify
Command: npx skills add https://github.com/Balthael/ciberbal-ai --skill sdd-verify-balthael

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After a change is implemented, teams need objective proof that the code actually matches the specification — not just that it exists. This Skill acts as a verification quality gate, executing tests, builds, and coverage checks to confirm behavioral compliance with specs before a change is archived. ## Core Features & Use Cases - Spec Compliance Matrix: Cross-references every spec scenario against actual test run results, classifying each as COMPLIANT, FAILING, UNTESTED, or PARTIAL. - Real Execution Verification: Detects and runs the project's test runner, build command, and coverage tool rather than relying on static analysis alone. - Strict TDD Mode: Optionally loads a strict-tdd-verify module that audits TDD cycle evidence, test layer distribution, per-file coverage, and assertion quality (detecting tautologies, ghost loops, and mock-heavy tests). - Use Case: An orchestrator completes an implementation phase and launches this Skill to verify the change; it runs the test suite, builds a compliance matrix, and returns a PASS / PASS WITH WARNINGS / FAIL verdict with CRITICAL, WARNING, and SUGGESTION issues. ## Quick Start Ask the agent to verify the completed change against its specs and tasks, executing the test suite and producing a verification report with a compliance matrix and verdict.

Frequently Asked Questions about sdd-verify

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

FAQPage Schema
How do I verify an implementation matches its specification?

Launch the verification phase after implementation completes. The Skill reads the specs, design, and tasks artifacts, executes the project's test suite and build, then maps every spec scenario to actual test results in a compliance matrix with a final PASS or FAIL verdict.

How does spec compliance verification work with test results?

Each spec scenario is matched to tests by name, description, or file path, then checked against the actual test run output. A scenario is only COMPLIANT when a covering test passed; missing tests are flagged UNTESTED and failing tests as CRITICAL.

What is Strict TDD mode in verification?

Strict TDD mode activates when strict_tdd is enabled and a test runner exists. It loads an additional module that audits the TDD cycle evidence from the apply phase, classifies tests by layer, reports per-file coverage for changed files, and scans for trivial assertions.

Does the verification skill fix the issues it finds?

No. The Skill only reports issues classified as CRITICAL, WARNING, or SUGGESTION. The orchestrator decides what action to take; the verifier never modifies code.

What happens if no coverage or quality tools are available?

Coverage and quality metric steps are skipped cleanly and reported as not available. Missing tools are never flagged as failures; only configured thresholds or detected problems produce warnings.

Why does verification require running tests instead of static analysis?

Static analysis only proves code exists, not that it behaves correctly. The Skill treats a spec scenario as compliant only when a test covering it has actually passed at runtime, making execution evidence mandatory.