vsdd

Enforce a spec-first workflow with TDD and adversarial verification.

2|Updated Feb 23, 2026

Search Tags:#tdd #spec-driven #verification #traceability #mutation-testing #ai-engineering #adversarial-review

Ready to run

npx skills add https://github.com/jonmumm/skills --skill vsdd

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: vsdd
Source: https://github.com/jonmumm/skills/tree/main/vsdd
Command: npx skills add https://github.com/jonmumm/skills --skill vsdd

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Verified Spec-Driven Development (VSDD) provides a disciplined, auditable lifecycle that ties a formal spec to tests and verification, reducing ambiguity and drift in AI-centric projects.

Core Features & Use Cases

Integrates spec crystallization, TDD implementation, adversarial roast, and formal hardening into a single, repeatable pipeline.
Maintains artifacts like vsdd/spec.md, vsdd/review-log.md, and vsdd/status.md for traceability and governance.
Supports adversarial reviews with structured feedback loops to converge toward zero-slop verification for high-stakes deployments.

Quick Start

Create a vsdd directory, then load references/phase1-spec.md to start Phase 1.

Frequently Asked Questions about vsdd

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

FAQPage Schema

What is spec-driven development for AI engineering projects?▼

Spec-driven development for AI engineering ties a formal specification directly to tests and verification, reducing ambiguity and drift throughout the project lifecycle. It enforces a rigorous pipeline from contract definition to validation.

How do I start a spec-driven workflow with traceability and governance?▼

To start a spec-driven workflow, create a vsdd directory and load the phase1 specification reference. This initiates the phased progression by crystallizing the contract definition into a formal spec.

How does adversarial review and mutation testing improve AI project verification?▼

Adversarial review and mutation testing improve AI project verification by applying structured feedback loops and static analysis to prove design correctness. This converges toward zero-slop verification for high-stakes deployments.

Can I use spec-driven development for formal verification and auditable governance?▼

Yes, spec-driven development supports formal verification and auditable governance. It maintains artifacts like status and review logs to ensure traceability across phased progression from contract definition to implementation.

What's the best way to prevent specification drift in test-driven development?▼

The best way to prevent specification drift in test-driven development is integrating spec crystallization with adversarial verification. Maintaining structured artifacts provides governance and traceability to enforce the formal spec.