impl

Coordinate Implementer, Reviewer, and Tester agents with worktree isolation.

7|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/Davie521/claude-skills --skill impl-davie521
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: impl
Source: https://github.com/Davie521/claude-skills/tree/main/plugins/dev-workflow/skills/impl
Command: npx skills add https://github.com/Davie521/claude-skills --skill impl-davie521

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates end-to-end multi-agent development workflows by coordinating Implementer, Reviewer, and Tester with worktree isolation to accelerate coding plans.

Core Features & Use Cases

  • Orchestrates a team of agents: Implementer, Reviewer, and Tester in a loop with immediate in-line reviews
  • Ensures worktree isolation for safe concurrent work and clean merge readiness
  • Supports plan-mode iterations (2–3 cycles) before implementation to minimize scope creep
  • Automatically triggers reviews and rework until all checks pass

Quick Start

Plan and execute a coding project using impl by outlining a plan, then start the Implementer to generate changes in an isolated worktree until reviews pass.

Frequently Asked Questions about impl

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

FAQPage Schema
How do I automate multi-agent coding workflows with isolated git worktrees?

Automate multi-agent coding by orchestrating Implementer, Reviewer, and Tester agents with isolated git worktrees. This ensures safe concurrent work, sequential agent handoffs, and clean merge readiness for review-ready commits.

What is the best way to coordinate code reviews and automated testing in a development workflow?

Coordinate code reviews and automated testing by triggering immediate in-line reviews and automated rework loops after implementation. This enforces sequential agent handoffs to automatically validate builds and deliver review-ready commits.

How does plan-mode iteration minimize scope creep before implementation?

Plan-mode iterations minimize scope creep by enforcing 2–3 planning cycles before implementation begins. This structured planning phase refines the coding plan to ensure scope alignment before agents generate code changes.

Can I use automated multi-agent development for projects requiring staged reviews?

Automated multi-agent development supports projects requiring staged reviews by orchestrating iterative planning, sequential agent handoffs, and automated validation. It triggers reviews and rework until all checks pass, delivering validated review-ready commits.

Does multi-agent code development work without manual intervention during the review process?

Multi-agent code development automates the review process by triggering reviews and rework loops until all checks pass. It coordinates agent handoffs automatically, requiring no manual intervention to validate builds and deliver review-ready commits.