Our take
Activerecent releases- Activity
- 57
- Community
- 49
- Issues
- 67
- Pull requests
- 67
+17,756 stars in the last 30 days
Superpowers offers a broad, structured workflow for coding agents and is actively maintained, though its development is concentrated among a small share of contributors.
Good fit if
- You use one of its supported coding-agent platforms and want a consistent process from design through implementation.
- You value explicit planning, test-driven development, and review steps in agent-led coding.
- You can work with its skill-based approach and, where relevant, Node.js project requirements.
Look elsewhere if
- You want an agent to start coding immediately without a requirements and planning workflow.
- Your team needs project development to be broadly distributed, since most contributions come from one person.
- You need a project with CI configured in the repository; the supplied project facts report tests but no CI.
All health signals
| Last commit | 2026-10-21 (7 days ago) |
|---|---|
| Commits, last 90 days | 6 |
| Releases, last 12 months | 16 (latest v7.0.1, 2026-10-21) |
| Contributors | 37 (top contributor: 80% of commits) |
| Issues closed, last 90 days | 48+ (typically closed in 7 days) |
| Pull requests merged, last 90 days | 19 (typically merged in 0 days) |
| Project age | 13 months |
Checked on 2026-10-29 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
319,438Forks
28,484Change since Nov 17, 2026: +692 stars·+79 forks
Overview
Superpowers is a software development methodology for coding agents, built on a composable library of skills and startup instructions that prompt agents to use them. It addresses the tendency of coding agents to jump straight into implementation by guiding them through clarifying intent, planning, test-driven development, and review.
The project supports a range of coding-agent platforms, including Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, and Pi. Its workflows are designed to work across supported agents rather than depend on one specific model or engine.
Key Features
- Brainstorming: elicits requirements and checks the agent’s understanding before work begins.
- Implementation planning: breaks approved designs into small tasks with file paths and verification steps.
- Test-driven development: promotes a red-green-refactor cycle during implementation.
- Subagent workflows: can delegate tasks to fresh agents and review their work, or execute plans inline.
- Git worktrees: sets up isolated branches and checks the project’s test baseline.
- Code review and branch completion: checks work against the plan and guides merge, pull request, or cleanup decisions.
- Session diagnosis: examines a session transcript for skill or workflow problems and can prepare a scrubbed report bundle.
Use Cases
- A developer wants an agent to clarify a feature’s requirements and present a design before changing code.
- A team wants agent-authored changes to follow a planned, test-first workflow with review checkpoints.
- A developer working on parallel tasks wants agents to use isolated Git worktrees and delegated subtasks.
- A coding-agent user needs to investigate why a skill did not trigger or a session deviated from its plan.
What you need
Detected in the repository
- Node.js (from package.json)
- A test suite
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
For Claude Code, install through the official plugin marketplace:
/plugin install superpowers@claude-plugins-official
Installation differs by coding-agent platform. See the README for the relevant instructions.
Alternatives
- agent-skills: It offers portable senior-engineer workflow skills across many agents, rather than a composable methodology spanning the full development process.
- skills: It provides individual engineering-practice skills without prescribing an end-to-end process framework.
- compound-engineering-plugin: It centers on a plugin workflow that captures learnings, while Superpowers provides a composable skills framework and development methodology.
- agent-skills: It emphasizes repeatable quality gates and autonomous sprint workflows, rather than Superpowers' broader skills-based methodology.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| superpowers | Shell | MIT | 308k | Active |
| agent-skills | JavaScript | MIT | 109k | Active |
| skills | Shell | MIT | 275k | Active |
| compound-engineering-plugin | TypeScript | MIT | 25.4k | Active |
| agent-skills | Shell | MIT | 2 | Active |
Considerations
The project is active, with recent commits and frequent releases, and its reported issue and pull-request handling is brisk. However, contributor activity is concentrated: one contributor accounts for most of it, which may make continuity more dependent on that person. The repository identifies Node.js as a runtime requirement, includes tests, and reports no CI configuration. Installation and update steps vary by coding-agent platform, so teams using multiple agents need to manage each integration separately. The project is MIT-licensed.
Found this useful?
Share it with someone who would like superpowers.