orchestrator-supaconductor

作者 Ibrahim-3d已验证

Multi-agent orchestration system for Claude Code with parallel execution, automated quality gates, Board of Directors, and bundled Superpowers skills

374
Stars
38
Forks
Python
语言
2026/8/23
添加时间

⚠️ 第三方软件声明

本 Skill 为第三方开源软件,独立托管于 GitHub。SkillTip 仅为信息目录,不控制或维护底层仓库。所显示的安全检查为自动化且范围有限,安装前请自行审查源码。

阅读服务条款

安装

添加到你的 Claude Code skills 目录:

# Add to your Claude Code skills
git clone https://github.com/Ibrahim-3d/orchestrator-supaconductor

快速入门

使用 orchestrator-supaconductor 等 Skills 的指南。

安全报告

已验证

上次扫描:—

{
  "status": "PASSED",
  "issues": []
}

README.md

SupaConductor

SupaConductor

Turn Claude Code into a full engineering team.
One command builds features, fixes bugs, and ships code — with automated planning, execution, and quality checks.

MIT License Version Claude Code Community

What Is This?InstallGet StartedHow It WorksCommandsFAQCommunity


What's New in v3.7.0

  • Board of Directors fast-path — Routine tracks now use a single Opus call (all 5 lenses in one pass) at ~1/10th the cost of the full multi-agent deliberation. Full board is reserved for high-stakes decisions: production deploys, security architecture, breaking API changes, data-loss migrations.
  • Plan revision guard — Prevents infinite planning loops. Tracks the number of times a plan is rejected and calls it done (with warnings) after 3 revisions.
  • Execution state reconciliation — On session resume, plan.md checkboxes are now treated as the source of truth. Prevents already-completed tasks from being re-executed after a crash.
  • Atomic file locks — Parallel workers can no longer accidentally acquire the same file lock simultaneously. Uses fcntl OS-level locking on a dedicated mutex file.
  • Bounded knowledge injection — The knowledge brief injected before planning is capped at 500 tokens (top-3 most relevant patterns and errors). Prevents context growth as your project's knowledge base accumulates.

What Does This Do?

You tell Claude Code what you want. SupaConductor figures out how to build it — step by step, with quality checks at every stage.

Without SupaConductor, you prompt Claude Code and hope for the best. You manually review, re-prompt, and fix what it misses.

With SupaConductor, you type one command:

/orchestrator-supaconductor:go Add user authentication with Google OAuth

And it automatically:

  1. Writes a detailed specification
  2. Creates an implementation plan with task dependencies
  3. Executes each task (in parallel when possible)
  4. Checks the work with specialized evaluators
  5. Fixes any issues it finds
  6. Marks the work complete when everything passes

No babysitting. No re-prompting. Just describe what you want and walk away.


What's Inside

ComponentCountWhat It Does
Commands36Slash commands you type to control everything
Skills39Specialized knowledge modules that activate when needed
Agents15Autonomous workers that handle planning, coding, and evaluation
Evaluators4Quality checkers for UI/UX, code quality, integrations, and business logic
Board of Directors5Virtual executives who deliberate on major decisions
Lead Engineers4Architecture, Product, Tech, and QA specialists

Bundles Superpowers v4.3.0 (MIT) — everything works out of the box with zero setup.


Installation

Option 1: Marketplace (recommended)

Open Claude Code and run:

/install Ibrahim-3d/orchestrator-supaconductor

Option 2: Clone from GitHub

git clone https://github.com/Ibrahim-3d/orchestrator-supaconductor.git ~/.claude/plugins/orchestrator-supaconductor

Option 3: Download manually

Download the latest release and extract it to ~/.claude/plugins/orchestrator-supaconductor/.

Verify it works

Start a new Claude Code session and type /orchestrator-supaconductor:. You should see a list of commands appear. If you see /orchestrator-supaconductor:go, you're all set.


Getting Started

Step 1: Set up your project

Open Claude Code in your project folder and run:

/orchestrator-supaconductor:setup

This walks you through an interactive setup that:

  • Analyzes your existing codebase (or starts fresh for new projects)
  • Helps define your product vision and tech stack
  • Creates a conductor/ folder in your project to track all work
  • Generates an initial development sprint with ready-to-execute tracks

You only need to do this once per project.

Step 2: Build something

Tell it what you want in plain English:

/orchestrator-supaconductor:go Add Stripe payment integration with webhooks
/orchestrator-supaconductor:go Fix the login bug where users get logged out after refresh
/orchestrator-supaconductor:go Build a dashboard with real-time analytics charts
/orchestrator-supaconductor:go Refactor the database layer to use connection pooling

That's it. SupaConductor takes over from here — planning, coding, testing, and evaluating the work automatically.

Step 3: Check on progress

/orchestrator-supaconductor:status

Shows all your active tracks, what step each one is on, and what's completed.

Step 4: Resume interrupted work

If a session ends before a track completes, just run:

/orchestrator-supaconductor:go

It picks up exactly where it left off.


How It Works

Every piece of work follows a structured cycle called the Evaluate-Loop:

Evaluate-Loop: Plan, Evaluate Plan, Execute, Evaluate Execution, Fix or Complete

StepWhat Happens
PlanBreaks your goal into tasks with a dependency graph
Evaluate PlanChecks for scope issues, overlap with other work, and feasibility
ExecuteWrites code, runs tests, tracks progress — parallel when possible
Evaluate ExecutionSpecialized checkers review UI/UX, code quality, integrations, and business logic
FixAddresses any failures, then loops back to evaluation (up to 5 fix cycles; up to 3 plan revisions)
CompleteAll checks pass — track is marked done

The loop runs fully automated. It stops when the work passes all quality checks or when it needs your input.

Parallel Execution

When tasks don't depend on each other, they run simultaneously:

Parallel execution showing independent tasks running at the same time

This means a feature with 6 tasks might only take as long as 3, because independent tasks run in parallel.

Two Modes

SupaConductor can work in two ways:

ModeBehaviorBest For
Agentic (default)Fully autonomous — makes all decisions itselfExperienced users who trust the system
Human-in-the-loopPauses at key decision points to ask youLearning the system, critical work, or when you want more control

Switch modes by editing conductor/config.json in your project:

{ "mode": "agentic" }
{ "mode": "human-in-the-loop" }

Board of Directors

For major decisions — and automatically for high-stakes tracks like production deploys, security architecture changes, breaking API migrations, or data-loss operations — a virtual board deliberates with written rationale:

Board of Directors: 5 executive perspectives deliberating on a decision

DirectorFocus
Chief ArchitectSystem design, scalability, technical debt
Chief Product OfficerUser value, market fit, feature scope
Chief Security OfficerVulnerabilities, compliance, data protection
Chief Operations OfficerFeasibility, timelines, deployment risks
Chief Experience OfficerUX/UI, accessibility, user journeys

Each director independently assesses your question, then they discuss and vote with written rationale.

/orchestrator-supaconductor:board-meeting Should we migrate from REST to GraphQL?
/orchestrator-supaconductor:board-review Add real-time notifications via WebSocket

Use board-meeting for full deliberation (detailed, takes longer) or board-review for quick assessments.


All Commands

The Main Ones

CommandWhat It Does
/orchestrator-supaconductor:go <goal>Describe what you want — everything else is automatic
/orchestrator-supaconductor:setupSet up SupaConductor in your project (run once)
/orchestrator-supaconductor:statusSee all your tracks and their progress
/orchestrator-supaconductor:implementContinue the Evaluate-Loop on the current track
/orchestrator-supaconductor:new-trackCreate a new track with more manual control

Quality and Review

CommandWhat It Does
/orchestrator-supaconductor:phase-reviewRun a quality gate on completed work
/orchestrator-supaconductor:cto-advisorGet a CTO-level architecture review
/orchestrator-supaconductor:board-meeting <topic>Full board deliberation with voting
/orchestrator-supaconductor:board-review <topic>Quick board assessment
/orchestrator-supaconductor:ui-auditAccessibility and UI/UX review

Expert Advisors

Ask for advice from virtual executives — they analyze your project and give guidance:

CommandAdvisor
/orchestrator-supaconductor:ceoBusiness strategy and product direction
/orchestrator-supaconductor:cmoMarketing strategy and positioning
/orchestrator-supaconductor:ctoTechnical architecture and engineering
/orchestrator-supaconductor:ux-designerUser experience and design

Planning and Execution

CommandWhat It Does
/orchestrator-supaconductor:writing-plansCreate a structured implementation plan
/orchestrator-supaconductor:executing-plansExecute an existing plan step by step
/orchestrator-supaconductor:brainstormingCreative exploration before building
/orchestrator-supaconductor:systematic-debuggingStructured approach to finding and fixing bugs
/orchestrator-supaconductor:using-git-worktreesIsolate feature work in separate git worktrees
/orchestrator-supaconductor:finishing-a-development-branchWrap up a branch — merge, PR, or cleanup

Loop Control (Advanced)

These give you fine-grained control over individual loop steps:

CommandStep
/orchestrator-supaconductor:loop-plannerRun just the planning step
/orchestrator-supaconductor:loop-plan-evaluatorEvaluate just the plan
/orchestrator-supaconductor:loop-executorRun just the execution step
/orchestrator-supaconductor:loop-execution-evaluatorEvaluate just the execution
/orchestrator-supaconductor:loop-fixerRun just the fix step
/orchestrator-supaconductor:parallel-dispatcherDispatch parallel workers manually
/orchestrator-supaconductor:task-workerRun a single task from the plan

Architecture

How the system fits together

                         /go <your goal>
                              |
                   +----------v-----------+
                   |    Orchestrator       |
                   |   (controls the loop) |
                   +----------+-----------+
                              |
          +-------------------+-------------------+
          v                   v                   v
     +--------+         +---------+         +----------+
     |  Plan  | ------> | Execute | ------> | Evaluate |
     +--------+         +---------+         +----------+
          |                   |                   |
          v                   v                   v
    writing-plans      parallel-dispatcher   4 evaluators
    plan-evaluator      |-- task-worker      |-- eval-ui-ux
    cto-reviewer        |-- task-worker      |-- eval-code-quality
                        +-- task-worker      |-- eval-integration
                                             +-- eval-business-logic

     +----------------------+    +----------------------------+
     |  Board of Directors  |    |  Knowledge / Retrospective |
     |  5 directors + vote  |    |  patterns.md + errors.json |
     +----------------------+    +----------------------------+

What gets created in your project

When you run /orchestrator-supaconductor:setup, it creates a conductor/ folder:

your-project/
+-- conductor/
    |-- tracks.md                # Registry of all your work tracks
    |-- config.json              # Mode setting (agentic or human-in-the-loop)
    |-- workflow.md              # How the development process works
    |-- authority-matrix.md      # Who can make which decisions
    |-- decision-log.md          # Record of architectural decisions
    |-- product.md               # Product vision and requirements
    |-- tech-stack.md            # Technology choices and constraints
    |-- knowledge/
    |   |-- patterns.md          # Patterns learned from completed work
    |   +-- errors.json          # Fixes for recurring errors
    +-- tracks/
        +-- feature-name/
            |-- spec.md          # What needs to be built
            |-- plan.md          # How to build it (tasks + dependencies)
            +-- metadata.json    # Current state and configuration

All of these are plain Markdown and JSON files. You can read, edit, or delete them anytime.


FAQ

Do I need to know how to code?

You need Claude Code installed and running — that requires basic terminal knowledge. But once SupaConductor is set up, you describe what you want in plain English and it handles the implementation.

How much does this cost in API credits?

SupaConductor uses the same Claude API as normal Claude Code — it just structures the work more carefully. Because it runs multiple agents (planning, execution, evaluation), it uses roughly 3-5x the API calls compared to doing everything manually in one conversation.

SupaConductor optimizes costs automatically: it uses Opus (the most capable model) for planning and evaluation, and Sonnet (faster, cheaper) for execution tasks.

Ways to reduce cost:

  • Use /orchestrator-supaconductor:implement if you write specs yourself
  • Skip board meetings for small features (they're opt-in)
  • Use human-in-the-loop mode to stay in control of scope

How much of my context window does this use?

Skills use progressive disclosure — only ~100 tokens each for metadata. Full instructions load only when activated (typically under 5,000 tokens each). The 39 skills are not loaded all at once.

Agents run as separate conversations with their own context windows, so they don't fill up your main conversation.

ComponentContext UsedWhen
Orchestrator~4,000 tokensActive during /go
Planner~3,000 tokensDuring planning only
Evaluator~2,500 tokens eachOnly the active evaluator loads
Board meeting~5,000 tokensOn-demand only
Idle~500 tokensBetween steps

Does this work with Cursor, Windsurf, or other AI tools?

No — SupaConductor is a Claude Code plugin that requires Claude Code's plugin system (agents, skills, slash commands, hooks).

However, the conductor/ directory it creates is just Markdown files. Any AI tool can read them. If you start with SupaConductor and switch tools later, your specs, plans, and documentation remain useful.

Is this overkill for small tasks?

For a quick one-file fix — yes, just use Claude Code directly. Use SupaConductor when:

  • The work touches 3 or more files
  • You'd normally plan before coding
  • You want automated quality checks
  • You want the work done right the first time

You can also use individual commands without the full loop:

/orchestrator-supaconductor:board-review Should we use Redis or PostgreSQL for sessions?
/orchestrator-supaconductor:cto-advisor
/orchestrator-supaconductor:writing-plans

Can I use this alongside other Claude Code plugins?

Yes. SupaConductor uses the /orchestrator-supaconductor: namespace and doesn't conflict with any other plugins, built-in commands, or MCP servers.

How do I update to a newer version?

If you installed via marketplace:

/install Ibrahim-3d/orchestrator-supaconductor

If you cloned via git:

cd ~/.claude/plugins/orchestrator-supaconductor && git pull

How do I uninstall?

# Disable without removing
/plugin    # Toggle it off in the plugin menu

# Full removal
rm -rf ~/.claude/plugins/orchestrator-supaconductor

The conductor/ directory in your project stays — it's just documentation files.


Migration from the Old Name

If you previously installed conductor-orchestrator-superpowers (the old name):

  1. Disable the old plugin: run /plugin in Claude Code and toggle it off
  2. Delete the old folder: rm -rf ~/.claude/plugins/conductor-orchestrator-superpowers
  3. Install SupaConductor using the instructions above
  4. Run /orchestrator-supaconductor:setup in your existing projects to update references

Your existing tracks and data are safe — only the command prefix changed from /conductor: to /orchestrator-supaconductor:.


Community

  • Discussions — Ask questions, share ideas, show what you've built
  • Issues — Report bugs or request features
  • Changelog — See what's new in each release

Requirements

Third-Party

Bundles Superpowers v4.3.0 by Jesse Vincent, licensed under MIT. See LICENSES/superpowers-MIT.

License

MIT — see LICENSE

常见问题

What is orchestrator-supaconductor?

orchestrator-supaconductor is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Ibrahim-3d. Multi-agent orchestration system for Claude Code with parallel execution, automated quality gates, Board of Directors, and bundled Superpowers skills. It has 374 GitHub stars.

Is orchestrator-supaconductor safe to use?

Yes. orchestrator-supaconductor passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.

How do I install orchestrator-supaconductor?

Clone the repository with "git clone https://github.com/Ibrahim-3d/orchestrator-supaconductor" and add it to your Claude Code skills directory (see the Installation section above).

What programming language is orchestrator-supaconductor written in?

orchestrator-supaconductor is primarily written in Python. It is open-source under Ibrahim-3d on GitHub, so you can review or fork the full source.

Are there alternatives to orchestrator-supaconductor?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh orchestrator-supaconductor against similar tools.

评论 (0)

暂无评论,成为第一个分享想法的人!

ECC

by affaan-m

10

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

242,21936,702JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情
15

An agentic skills framework & software development methodology that works.

234,96620,863Shell
AI 智能体ai-agentsbrainstorming
查看详情

hermes-agent

by NousResearch

10

The agent that grows with you

234,43747,175Python
AI 智能体ai-agentsagent-orchestration
查看详情

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

185,94028,768JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情

cc-switch

by farion1231

3

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

128,8688,826Rust
AI 智能体claude-codeai-tools
查看详情

claude-code

by anthropics

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

120,03119,897Shell
AI 智能体
查看详情

开发者还喜欢

基于喜欢此 Skill 的开发者投票和收藏

ECC

by affaan-m

10

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

242,21936,702JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情
15

An agentic skills framework & software development methodology that works.

234,96620,863Shell
AI 智能体ai-agentsbrainstorming
查看详情

hermes-agent

by NousResearch

10

The agent that grows with you

234,43747,175Python
AI 智能体ai-agentsagent-orchestration
查看详情

n8n

by n8n-io

12

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

201,88160,308TypeScript
MCP 服务器apisai-tools
查看详情

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

185,94028,768JavaScript
AI 智能体ai-agentsanthropicclaude-code
查看详情

cc-switch

by farion1231

3

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io

128,8688,826Rust
AI 智能体claude-codeai-tools
查看详情