Agent skill · Superpowers
/subagent-driven-development Subagent-driven plan execution with per-task review
Subagent-driven development executes an implementation plan by giving each task to a fresh subagent with only the context it needs, then reviewing the result for spec compliance and code quality. Progress is tracked in a ledger file so the work survives context compaction, and a whole-branch review runs at the end.
Works with
- Claude Code
- Codex CLI
- Cursor
- Gemini CLI
- GitHub Copilot CLI
- Devin CLI
- Qwen Code
- Pi
- Antigravity
Used forSubagents and long tasks
When to use it
- You have an approved plan whose tasks are mostly independent
- Your agent supports subagents and you want thorough review after every task
- A long session needs to keep going for hours without drifting from the plan
How it works
- 1Set up the workspace
It confirms an isolated worktree, creates a per-plan ledger under .superpowers/sdd/ and scans the plan for conflicts.
- 2Dispatch an implementer
Each task goes to a new subagent that implements, tests, commits and self-reviews.
- 3Review the task
A reviewer subagent checks the change against the spec and for quality, with a fix loop when issues are found.
- 4Record and continue
Completion and any judgment calls are written to the ledger, then the next task starts without pausing.
- 5Final review
After all tasks, a broad review of the whole branch runs before handing off to finishing-a-development-branch.
What it steers the agent away from
- Re-running already finished tasks after the agent loses context
- Stopping to ask for permission between every task
- Starting implementation directly on the main branch without consent
Install this skill
It installs with the Superpowers pack.
Claude CodeFrom Anthropic's official plugin marketplace.
/plugin install superpowers@claude-plugins-officialClaude Code (Superpowers marketplace)Alternative marketplace that also lists related plugins.
/plugin marketplace add obra/superpowers-marketplace
/plugin install superpowers@superpowers-marketplaceCodex CLISearch for superpowers in the plugin list and choose Install Plugin; the Codex app has the same entry under Plugins.
/pluginsCursorRun in Cursor Agent chat, or search the plugin marketplace.
/add-plugin superpowersGemini CLIUpdate later with gemini extensions update superpowers.
gemini extensions install https://github.com/obra/superpowersGitHub Copilot CLIRegisters the marketplace, then installs the plugin.
copilot plugin marketplace add obra/superpowers-marketplace
copilot plugin install superpowers@superpowers-marketplaceDevin CLIUpdate with devin plugins update superpowers.
devin plugins install obra/superpowersQwen CodePick superpowers when prompted.
qwen extensions install obra/superpowersPiLoads the skills plus a small extension that injects the bootstrap at session start.
pi install git:github.com/obra/superpowersAntigravityRun the same command again to update.
agy plugin install https://github.com/obra/superpowersCommands come from the author’s README; check it for changes before you run them.
Questions
How is it different from executing-plans?
Executing-plans does every task inline in one context with a single review at the end, which costs less; subagent-driven development uses a fresh context per task and reviews each one, which is more thorough.
Which models does it use?
It recommends the least capable model that can handle each role: cheaper models for mechanical tasks and the most capable one for design work and the final review.
More guides from Superpowers
/brainstormingDesign-first brainstorming before any buildClarifies what you want and why, then gets a design approved before building.Read the guide →/test-driven-developmentStrict red-green-refactor test-driven developmentEnforces writing a failing test first, then minimal code, then cleanup.Read the guide →
Also in this pack
/diagnosing-superpowersAnalyzes a session transcript to explain what went wrong, with optional bug report.↗/dispatching-parallel-agentsSplits independent tasks across several subagents that work at the same time.↗/receiving-code-reviewEvaluates review feedback critically and verifies it before changing any code.↗/requesting-code-reviewAsks a reviewer to check finished work against the plan and requirements.↗/systematic-debuggingFinds the root cause of a bug in four phases before attempting a fix.↗/using-superpowersTeaches the agent to look for and load matching skills before responding.↗/verification-before-completionRequires running checks and reading their output before claiming work is done.↗/writing-skillsGuides creating, editing and testing new skills before they are used.↗
Summary by trendhub. The skill itself lives in obra/superpowers/skills/subagent-driven-development by Jesse Vincent (MIT); read the full, current text there.