askill
refactoring-team

refactoring-teamSafety 78Repository

Iterative code refactoring through progressive lenses via a worker-reviewer agent team.

193 stars
3.9k downloads
Updated 3/25/2026

Package Files

Loading files...
SKILL.md

STARTER_CHARACTER = πŸ’Ž

Prerequisites

Agent teams must be enabled in settings:

{
  "env": {
    "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1"
  }
}

If not set, offer to add it before proceeding.

Setup

If $ARGUMENTS provided, use as target path. Otherwise ask for:

  • Target path (files or folder to refactor)
  • Test command to verify changes

Verify the target path exists and tests pass before proceeding.

Launch the Team

Generate a short random ID: head -c 3 /dev/urandom | xxd -p | head -c 3

Use it to name the teammates:

  • Worker: worker-ID (e.g. worker-a3f)
  • Reviewer: reviewer-ID (e.g. reviewer-a3f)

Read the spawn prompts:

Before spawning, replace these placeholders in both prompts:

  • TARGET_PATH β†’ actual target path
  • TEST_COMMAND β†’ actual test command
  • LENSES_DIR β†’ ${CLAUDE_SKILL_DIR}/references/lenses
  • GUIDES_DIR β†’ ${CLAUDE_SKILL_DIR}/references/reviewer-guides
  • WORKER_NAME β†’ the worker's name (e.g. worker-a3f)
  • REVIEWER_NAME β†’ the reviewer's name (e.g. reviewer-a3f)

After Launch

Tell the user:

  • Shift+Down cycles between worker and reviewer
  • For split panes: set teammateMode: "tmux" in settings

Monitor Progress

When a worker goes idle, read .refactoring-state to check the lens number. If it skipped a value, message the reviewer to go back and complete the skipped lens first.

Install

Download ZIP
Requires askill CLI v1.0+β–Ά

AI Quality Score

68/100Analyzed 3/30/2026

A well-structured, highly actionable skill for running a worker-reviewer agent team for code refactoring. The skill is clear and organized with proper CLI integration, safety hooks, and clear sequential steps. Score penalized by heavy reliance on external reference files (worker-prompt.md, reviewer-prompt.md, lenses, guides) that aren't provided, making standalone execution uncertain. Claude Code-specific implementation but the workflow pattern is broadly applicable.

78
82
72
65
75

Metadata

Licenseunknown
Version-
Updated3/25/2026
Publisherlexler

Tags

promptingtesting