superpowers: Guide Coding Agents Through Software Development

superpowers: Guide Coding Agents Through Software Development

Summary

Superpowers adds a structured development methodology to coding agents, from clarifying requirements and planning through test-driven implementation and review. It suits developers who want repeatable workflows across supported agent tools.

At a glance

Language
Shell
License
MIT
Stars
295k
Forks
26.3k
Added to OSRepos
June 12, 2026
Last analyzed
October 3, 2026
View on GitHub

Topics

Click on any tag to explore related repositories

Use at your own risk

OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.

Overview

Superpowers is a framework of reusable skills and workflow instructions for coding agents. It addresses a common gap in agent-assisted development: agents may start coding before requirements are clear, skip tests, or drift from a plan.

Instead, it guides work through design, planning, implementation, review, and branch completion. It is intended for developers who want a consistent, process-driven approach across supported coding-agent tools, rather than a standalone coding agent or general-purpose application.

Key Features

  • Brainstorming skill to clarify requirements, explore alternatives, and validate a design before implementation.
  • Planning skill that breaks approved designs into small tasks with file paths and verification steps.
  • Test-driven development workflow based on a red, green, refactor cycle.
  • Task execution through subagents with reviews, or inline execution with a final review.
  • Skills for systematic debugging, code review, worktrees, and finishing development branches.
  • Integrations for multiple coding-agent harnesses, with separate installation for each.
  • Session diagnosis skill that can inspect a transcript and prepare a scrubbed report for a bug report.

Use Cases

  • A solo developer wants an agent to clarify an early-stage idea and produce a reviewable plan before changing code.
  • A team using coding agents wants to encourage test-first implementation and consistent review between tasks.
  • A developer tackling a multi-step feature wants an agent to execute a plan using subagents or work in an isolated branch.
  • An agent user needs to investigate why a workflow misfired or deviated during a session.

Project Facts

  • Language: Shell
  • License: MIT
  • Stars: 295k
  • Forks: 26.3k
  • Topics: ai, brainstorming, coding, obra, sdlc, skills, subagent-driven-development, superpowers
  • Archived: No

Getting Started

Install Superpowers using the method for your coding-agent harness. For Claude Code, the README lists this command:

/plugin install superpowers@claude-plugins-official

See the README for other integrations, setup details, and update instructions.

Alternatives

  • agent-skills: It offers portable engineering workflows, while Superpowers packages a structured methodology for supported agent tools.
  • skills: It provides composable skills for specific practices rather than Superpowers' end-to-end development workflow.
  • compound-engineering-plugin: It guides work through brainstorming, implementation, review, and captured learnings, with support for several agent hosts.

Considerations

  • Superpowers provides workflows and skills, not a coding model or standalone agent. You need a compatible coding-agent harness and must install it for each harness you use.
  • Its workflow is deliberately prescriptive, including mandatory skills and test-driven development. That can add process overhead when a task is small or exploratory.
  • The README says contributions of new skills are generally not accepted, and skill changes must work across supported agents.
  • Optional visual-companion telemetry loads a versioned logo from the project's website. The README says it does not include project or prompt details and can be disabled with SUPERPOWERS_DISABLE_TELEMETRY.

Source repository

Open the original repository on GitHub.

26 counted GitHub visits

View on GitHub

Related repositories

Similar repositories that may be relevant next.

web-design: A Claude Code SKILL for Spec-First Web Page Design

web-design: A Claude Code SKILL for Spec-First Web Page Design

October 3, 2026

The web-design project is a Claude Code SKILL designed to streamline the creation of beautiful and consistent web pages. It emphasizes a 'spec first, code second' approach, ensuring design principles are established before development begins. This tool helps generate UI, visuals, motion, and responsiveness that are consistent across pages and easily editable.

Claude CodeClaude SkillDesign System
OrcaReplay: Time Travel for AI Agents, Debugging and Evaluation

OrcaReplay: Time Travel for AI Agents, Debugging and Evaluation

October 2, 2026

OrcaReplay introduces "time travel" capabilities for AI agents, allowing developers to record, replay, fork, and debug any agent run with any model. It addresses the challenges of AI agent debugging by providing byte-for-byte reproducibility, offline analysis, and the ability to compare different models from specific checkpoints. This tool, built by the OrcaRouter.ai team, enhances observability and control over complex agent behaviors.

Agent DebuggingAI AgentsLLM Agents
Shepherd: Reversible Execution Traces for Programmable Meta-Agents

Shepherd: Reversible Execution Traces for Programmable Meta-Agents

October 2, 2026

Shepherd is a Python runtime substrate designed for agent work requiring inspection, reversibility, and supervision. It records agent runs as durable, inspectable execution traces, enabling meta-agents to observe, fork, replay, and revert any operation. This framework couples agents and environments using a copy-on-write fork, offering significant performance benefits and robust permission enforcement.

PythonAIAgent Framework
SwarmLLM: Run Local AI Models and Team Up for Giant Distributed Inference

SwarmLLM: Run Local AI Models and Team Up for Giant Distributed Inference

October 2, 2026

SwarmLLM is a free, open-source application that allows you to run AI chat models directly on your own computer. It uniquely enables multiple computers to team up over the internet, collectively running models too large for a single machine. This platform offers an OpenAI and Anthropic-compatible API, all without requiring accounts or cryptocurrency.

AILLMDecentralized
OS
OSRepos

Analysis and discovery of open source repositories. Find interesting projects and follow their updates.

Monitor your website with YourWebsiteScore

OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of third-party repository code is at your own risk. Always review source code, dependencies, licenses, and security implications before running anything.

© 2025 OSRepos. Built with Nuxt 3 and lots of ❤️