code-conductor

🥈Silver

Code Conductor enables operations teams to run multiple Claude Code sub agents in parallel, accelerating development cycles. It provides GitHub-native orchestration, eliminating conflicts and complexity. Teams can ship features 10x faster by leveraging AI coding agents for rapid, conflict-free development.

9070Updated 1mo ago

Intermediate30min to implementautomation

Saves ~240 min per use

Quick InstallView Source

git clone https://github.com/ryanmac/code-conductor.git

Works with:

Claude

Overview

About This Skill

How to Use

[{"step":"Define the scope and components. Break down your feature into 3-5 distinct components (e.g., frontend, backend API, database schema, tests, documentation). List them in order of dependency.","tip":"Use the prompt template to specify components clearly. Avoid overlapping responsibilities between agents to prevent conflicts."},{"step":"Initialize the parallel agents. Run the prompt in your AI tool, replacing placeholders with your feature name, project name, and components. Ensure you have write access to the GitHub repository.","tip":"Use Claude Code's `gh` CLI integration for seamless GitHub operations. Verify all agents have the same base branch to avoid merge hell later."},{"step":"Monitor progress and resolve conflicts. Check in every 20-30 minutes. Use `git status` and GitHub PRs to track changes. Intervene immediately if conflicts arise (e.g., same file modified by two agents).","tip":"Set up GitHub notifications for PRs and CI status. Use `gh pr view` to inspect changes quickly."},{"step":"Merge and validate. Once all agents complete their PRs, merge them into a feature branch. Run the full test suite and review the deploy preview. Only merge into `main` after approval.","tip":"Use `gh pr merge --auto --squash` for clean history. Validate the deploy preview URL thoroughly before proceeding."},{"step":"Document and deploy. Update any remaining documentation (e.g., README, API docs) and trigger the deployment pipeline. Share the success metrics (e.g., time saved, lines of code) with the team.","tip":"Log the start/end times and agent outputs for future reference. Compare against manual estimates to showcase efficiency gains."}]

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/ryanmac/code-conductor

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Act as Code Conductor. Use [NUMBER] parallel Claude Code sub-agents to implement [FEATURE_NAME] for [PROJECT_NAME]. Each agent should handle a distinct component: [COMPONENT_1], [COMPONENT_2], and [COMPONENT_3]. Coordinate their work to avoid conflicts, merge changes via GitHub PRs, and ensure all tests pass. Report progress every 30 minutes and resolve any merge conflicts immediately. Start by creating a GitHub issue with the task breakdown and assign agents accordingly.

Example Output

### Code Conductor Session Log - Feature: User Profile Dashboard Implementation

**Initialization (08:45 AM)**
- Triggered 3 parallel sub-agents via Claude Code:
  - Agent A (Frontend): React component for profile card with responsive design
  - Agent B (Backend): REST endpoint for profile data with caching layer
  - Agent C (Database): Schema migration for new user preferences table
- Created GitHub issue #423 with task breakdown and assigned agents
- Generated branch `feature/user-profile-dashboard` from `main`

**Progress Report (09:15 AM)**
- Agent A completed: ProfileCard.jsx (80% test coverage), responsive CSS grid layout
- Agent B completed: GET `/api/profile/{id}` endpoint with Redis caching (latency < 50ms)
- Agent C encountered conflict: Existing `users` table migration detected. Auto-resolved by renaming to `user_preferences` and added foreign key constraint.
- All agents pushed changes to their respective sub-branches:
  - `agent-a/profile-card`
  - `agent-b/profile-api`
  - `agent-c/db-migration`

**Conflict Resolution (09:22 AM)**
- Detected merge conflict in `package.json` (Agent A added Tailwind, Agent B needed older version for backend). Auto-resolved by pinning Tailwind to v3.3.0 in both branches.
- Triggered CI pipeline for all three branches. All passed except Agent B's backend tests (mock database timeout). Agent B rerouted to use staging DB temporarily.

**Final Merge (09:45 AM)**
- Agent A's PR #424 merged into `feature/user-profile-dashboard` (12 files, 450 lines added)
- Agent B's PR #425 merged after fixing timeout issue (8 files, 210 lines added)
- Agent C's PR #426 merged with schema changes (3 files, 150 lines added)
- Created PR #427 to merge `feature/user-profile-dashboard` into `main`
- All tests passed (frontend: 95% coverage, backend: 92% coverage)
- Deploy preview generated: https://pr-427.preview.company.com

**Completion (09:50 AM)**
- Feature implemented in 65 minutes (vs. estimated 8 hours manually)
- 3 parallel agents worked on distinct components without conflicts
- GitHub-native orchestration handled branch management and CI/CD automatically
- Ready for code review and production deployment.