Question 1

What Does This Do?

Accepted Answer

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, manually reviewing, re-prompting, and fixing what it misses. With SupaConductor, you type one command 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.

Question 2

Can I use this alongside other Claude Code plugins?

Accepted Answer

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

Question 3

Is this overkill for small tasks?

Accepted Answer

For a quick one‑file fix, yes — just use Claude Code directly. Use SupaConductor when the work touches three or more files, you would normally plan before coding, you want automated quality checks, or you want the work done right the first time. You can also use individual commands without the full loop.

Question 4

Cursor、Windsurf、または他のAIツールと互換性がありますか？

Accepted Answer

いいえ — SupaConductorはClaude Codeのプラグインシステム（エージェント、スキル、スラッシュコマンド、フック）を必要とするClaude Codeプラグインです。ただし、生成されるconductor/ディレクトリはMarkdownファイルだけです。任意のAIツールでこれらを読むことができるため、後でツールを切り替えても、仕様、計画、ドキュメントを保持できます。

Question 5

How much of my context window does this use?

Accepted Answer

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. | Component | Context Used | When | |-----------|-------------|------| | Orchestrator | ~4,000 tokens | Active during /go | | Planner | ~3,000 tokens | During planning only | | Evaluator | ~2,500 tokens each | Only the active evaluator loads | | Board meeting | ~5,000 tokens | On‑demand only | | Idle | ~500 tokens | Between steps |

Question 6

How much does this cost in API credits?

Accepted Answer

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 3-5× 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 include using /orchestrator-supaconductor:implement if you write specs yourself, skipping board meetings for small features (they're opt‑in), and using human‑in‑the‑loop mode to stay in control of scope.

Question 7

Do I need to know how to code?

Accepted Answer

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

Question 8

What gets created in your project?

Accepted Answer

When you run /orchestrator-supaconductor:setup, it creates a conductor/ folder containing plain Markdown and JSON files. You can read, edit, or delete them anytime.

Question 9

How It Works?

Accepted Answer

Every piece of work follows a structured cycle called the Evaluate-Loop: | Step | What Happens | |------|--------------| | Plan | Breaks your goal into tasks with a dependency graph | | Evaluate Plan | Checks for scope issues, overlap with other work, and feasibility | | Execute | Writes code, runs tests, tracks progress — parallel when possible | | Evaluate Execution | Specialized checkers review UI/UX, code quality, integrations, and business logic | | Fix | Addresses any failures, then loops back to evaluation (up to 5 fix cycles; up to 3 plan revisions) | | Complete | All checks pass — track is marked done | The loop runs fully automated and stops when the work passes all quality checks or when it needs your input.

Question 10

What's Inside?

Accepted Answer

| Component | Count | What It Does | |-----------|------:|-------------| | Commands | 36 | Slash commands you type to control everything | | Skills | 39 | Specialized knowledge modules that activate when needed | | Agents | 15 | Autonomous workers that handle planning, coding, and evaluation | | Evaluators | 4 | Quality checkers for UI/UX, code quality, integrations, and business logic | | Board of Directors | 5 | Virtual executives who deliberate on major decisions | | Lead Engineers | 4 | Architecture, Product, Tech, and QA specialists | Bundled as Superpowers v4.3.0 (MIT) — everything works out of the box with zero setup.

Question 11

新しいバージョンにアップデートするにはどうすればよいですか？

Accepted Answer

マーケットプレイスからインストールした場合：
Gitでクローンした場合：

Question 12

What's New in v3.7.0?

Accepted Answer

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.

Question 13

What are the limitations of using Git Worktrees?

Accepted Answer

It may require manual intervention if tests fail during baseline verification and assumes a specific directory structure and gitignore configuration.

Question 14

What are the key benefits of using Git Worktrees?

Accepted Answer

It enables work on multiple branches simultaneously, prevents accidental commits, and automates directory selection and safety verification.

Question 15

How does directory selection work?

Accepted Answer

The directory selection process checks for existing directories with priority order, then checks for a preference in the 'CLAUDE.md' file, and uses that if found.

Question 16

What is the purpose of Git Worktrees?

Accepted Answer

Git Worktrees creates isolated git worktrees with smart directory selection and safety verification for feature work that needs isolation from the current workspace.

Using Git Worktrees (Grade A)セキュリティテスト済みデータAISHILL(Grade A)。クリュードAIと併用する。

概要

主な機能

料金

ユースケース

隔離された機能開発

実装計画

メリット & デメリット

メリット

デメリット

レビュー

Q&A

質問する

東学サェソーの代替

Manage Headers (Grade A)

Ga4 Bigquery Schema (Grade A)

Meta Capi (Grade A)

Callees (Grade A)

Test Module Name (Grade A)

Board Of Directors (Grade A)

Advpl Mvc Avancado (Grade A)

Docs Writer (Grade A)

概要