Superpowers
Superpowers is an open-source agentic skills framework that gives Claude Code and other coding agents structured software development methodologies.
Overview
Superpowers helps developers improve AI-assisted coding by adding reusable software development skills, structured workflows, test-first practices, systematic debugging, brainstorming, subagent delegation, code review habits, and custom skill creation to agentic coding tools.
Core Features & Capabilities
Ideal for software developers, AI-assisted coding users, Claude Code users, Codex users, technical founders, engineering teams, open-source maintainers, coding-agent builders, prompt engineers, DevOps engineers, QA engineers, and developers who want more reliable AI coding workflows.
- Add structured software development skills to Claude Code and other AI coding agents
- Use test-driven development workflows to guide agents toward safer implementation
- Apply systematic debugging methods instead of letting agents guess at fixes
- Use brainstorming, planning, subagent workflows, and code review to improve AI-generated code quality
- Author new custom skills so agents can follow repeatable team-specific engineering practices
Trending Use Cases
Why Developers Choose Superpowers
Visit the Superpowers Claude plugin page or GitHub repository, then install it through the official Claude Code plugin marketplace or the supported setup method for your coding environment. For Claude Code, install the plugin from the official marketplace, then begin using it during software development tasks where planning, tests, debugging, subagents, or code review would improve reliability. Developers can also explore the repository documentation to install Superpowers for Codex, OpenCode, Cursor, Gemini CLI, GitHub Copilot CLI, and other supported agentic coding tools.
“Superpowers helps coding agents follow better engineering discipline by giving them reusable skills for planning, testing, debugging, review, and development workflows.”
Getting Started with Superpowers
By combining reusable agentic skills, TDD workflows, systematic debugging, brainstorming, subagent-driven development, code review, custom skill authoring, and multi-agent-tool support, Superpowers gives developers a more disciplined way to use AI coding agents for real software engineering work.
Open the tool and review its core product experience.
Create your account or access your existing workspace.
Use your own task to judge speed, quality, and fit.
Check similar AI tools before making a final decision.
Superpowers
Superpowers is an open-source agentic skills framework created by Jesse Vincent / obra for Claude Code and other AI coding agents, focused on structured software development, TDD, debugging, planning, subagents, code review, and reusable skills.
Superpowers is an open-source plugin and agentic skills framework designed to make AI coding agents behave more like disciplined software development partners. Instead of relying only on raw prompt instructions, Superpowers provides composable skills and workflows for test-driven development, systematic debugging, planning, brainstorming, subagent-driven development, code review, skill authoring, and repeatable engineering practices. It is available through the official Claude plugin marketplace and can also be installed across several coding-agent environments, including Claude Code, Codex CLI, Codex App, OpenCode, Cursor, Gemini CLI, Factory Droid, and GitHub Copilot CLI. Superpowers is especially useful for developers who want AI agents to slow down, plan properly, write tests first, review work, isolate tasks, and follow more reliable software delivery patterns.
github.com/obra/superpowers
Comments (0)
No Comments Found