phase-gate

Evaluate implementation phase readiness by analyzing scope completion and running quality gates.

Updated Jul 9, 2026
One-click install
npx skills add https://github.com/ByronWilliamsCPA/plugin --skill phase-gate-byronwilliamscpa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: phase-gate
Source: https://github.com/ByronWilliamsCPA/plugin/tree/main/plugins/wff-code/skills/phase-gate
Command: npx skills add https://github.com/ByronWilliamsCPA/plugin --skill phase-gate-byronwilliamscpa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Determining whether an implementation phase is truly complete requires checking both deliverable scope and code quality, which is error-prone when done manually. This Skill automates that evaluation so phases only transition when verified criteria are met. ## Core Features & Use Cases - Parallel Scope and Quality Analysis: Dispatches scope-analyzer and phase-reviewer agents simultaneously to check deliverables against the implementation plan and run lint, type, test, coverage, and security gates. - Multiple Evaluation Modes: Run full evaluations, scope-only progress checks, review-only quality verification, or pre-implementation plan validation with worktree setup. - Approval-Gated Transitions: Presents a synthesized verdict and blocks advancement to the next phase until the user explicitly approves. - Use Case: Before merging a phase branch, run a full gate review to confirm all deliverables are done, tests pass with adequate coverage, and no scope creep exists, then get guided through merge and next-phase setup. ## Quick Start Ask the AI to run a phase gate evaluation for phase 1 to check whether it is complete and ready for the next phase.

Frequently Asked Questions about phase-gate

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

FAQPage Schema
How do I check if an implementation phase is complete?▼

Run a full phase gate evaluation with the phase number, which dispatches scope and quality review agents in parallel. It returns a verdict of READY or NOT READY with specific blockers, and waits for your approval before advancing.

How to validate an implementation plan before writing code?▼

Use the plan mode, which locates the phase plan file, runs a plan-validator agent to check traceability against acceptance criteria, flags scope creep, and sets up an isolated git worktree after your approval.

What quality gates does a phase review run?▼

The phase-reviewer agent runs ruff check, ruff format --check, basedpyright type checking, pytest with coverage measurement, bandit security scanning, and the per-phase smoke test defined in the implementation plan.

Can I check phase progress without running the full test suite?▼

Yes, use scope-only mode to run just the scope-analyzer agent, which reports deliverable status as DONE, PARTIAL, NOT_STARTED, or UNCLEAR plus any scope creep, without executing quality gates.

What happens when a phase gate verdict is NOT READY?▼

The skill lists the specific blockers from both agent reports and offers to help fix them. It stops immediately and never begins work on the next phase until you explicitly approve after the blockers are resolved.