sdd-verify

Evaluate SDD implementations against spec acceptance criteria and produce a structured PASS/FAIL verdict.

Updated Apr 20, 2026
One-click install
npx skills add https://github.com/juanmanueldaza/invencible --skill sdd-verify-juanmanueldaza
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-verify
Source: https://github.com/juanmanueldaza/invencible/tree/main/.claude/skills/sdd-verify
Command: npx skills add https://github.com/juanmanueldaza/invencible --skill sdd-verify-juanmanueldaza

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Determines whether an SDD implementation satisfies all acceptance criteria from the spec and returns a clear, actionable verdict.

Core Features & Use Cases

  • Reads the spec and the implementation from memory and evaluates each acceptance criterion.
  • Produces a structured PASS/FAIL/PARTIAL verdict with evidence for traceability.
  • Saves the verdict back into memory to support audits, reviews, and governance.

Quick Start

Run sdd-verify after applying changes to automatically produce a structured 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 if my implementation satisfies acceptance criteria from a spec?

To verify acceptance criteria, the implementation and spec are read from memory, evaluated against each requirement, and a structured PASS or FAIL verdict with evidence is produced for traceability.

What is a structured verdict for SDD implementations?

A structured verdict is an automated evaluation result that checks an implementation against spec acceptance criteria and outputs a PASS, FAIL, or PARTIAL status with supporting evidence saved back to memory.

How do I run automated acceptance criteria checks after applying code changes?

Automated acceptance criteria checks run after applying changes by reading the spec and implementation from memory, evaluating each criterion, performing optional tests, and saving a structured verdict.

Can I produce a PASS/FAIL status for governance and audit reviews?

Yes, a PASS/FAIL status is produced for governance and audit reviews by evaluating the implementation against the spec and saving the structured verdict with traceable evidence back into memory.

What are the limitations of automated spec verification?

Automated spec verification requires the spec and implementation to be pre-loaded in memory and applies after changes are made, meaning it cannot evaluate unimplemented criteria or external dependencies outside the memory context.

Does spec verification work without writing manual tests?

Spec verification works without manual tests by reading the spec and evaluating the implementation directly from memory, though it can also perform optional tests to strengthen the final verdict.