A quality gate sub-agent that verifies code implementation against provided design specs and task lists through execution.

Install

mkdir -p .claude/skills/sdd-verify && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/15029" && unzip -o skill.zip -d .claude/skills/sdd-verify && rm skill.zip

Installs to .claude/skills/sdd-verify

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Validate that implementation matches specs, design, and tasks. Trigger: When the orchestrator launches you to verify a completed (or partially completed) change.
161 chars✓ has a “when” trigger
Intermediate

Key capabilities

  • →Verify task completion against `tasks.md`
  • →Check code correctness against spec requirements
  • →Validate design decisions against `design.md`
  • →Run project tests and capture results
  • →Execute build and type-check commands
  • →Generate a spec compliance matrix based on test results

How it works

This skill acts as a verification agent, systematically checking the completeness and correctness of an implementation against project specifications, design, and tasks. It involves static analysis, running tests, executing build commands, and generating a spec compliance matrix to provide evidence of behavioral compli

Inputs & outputs

You give it
Project documentation (spec, design, tasks) and codebase
You get back
A detailed verification report including task status, compliance matrix, test results, and identified issues

When to use sdd-verify

  • →Verify implementation against spec
  • →Check task completion status
  • →Validate code compliance

About sdd-verify

This skill acts as a verification sub-agent, checking completion of tasks and correctness of code. It requires executing the code to prove it meets the requirements outlined in the project documentation.

>

When not to use it

  • →When static analysis alone is considered sufficient for verification
  • →When the goal is to fix issues rather than report them
  • →When ignoring the project's existing test runner or build system

Limitations

  • →Does not fix any issues, only reports them
  • →Requires a test runner and build system to be detectable
  • →Coverage validation is only performed if a threshold is configured

How it compares

This skill provides a structured, execution-based verification process that goes beyond static analysis, generating a detailed compliance matrix by cross-referencing spec scenarios with actual test results.

Compared to similar skills

sdd-verify side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
sdd-verify (this skill)07moNo flagsIntermediate
python-testing-patterns776moReviewIntermediate
dependency-upgrade268moReviewIntermediate
test-cases5710moNo flagsBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry