Frequently Asked Questions

Superpowers is a complete software development methodology for your coding agents, built on top of a set of composable skills and initial instructions. It transforms how your agents approach coding by implementing structured workflows, test-driven development, and systematic debugging.

Superpowers activates automatically when your coding agent starts working. Instead of jumping directly into code writing, the agent first asks you what you're trying to accomplish. It then refines a specification, creates an implementation plan, and uses subagent-driven development to execute tasks systematically.

Superpowers supports multiple coding agents including Claude Code, Codex CLI, Codex App, Factory Droid, Gemini CLI, OpenCode, Cursor, and GitHub Copilot CLI. Each has its own installation process through their respective plugin marketplaces or extension systems.

Installation varies by harness. For Claude Code, use `/plugin install superpowers@claude-plugins-official`. For Codex CLI, search for "superpowers" in the plugin marketplace. For Gemini CLI, use `gemini extensions install https://github.com/obra/superpowers`. Each harness has its specific installation command as detailed in our installation guide.

Yes, you need to install Superpowers separately for each harness you use. Each coding agent has its own plugin system, so installation is specific to each platform. However, once installed, all harnesses will benefit from the same comprehensive skill set.

Yes, Superpowers is open source and free to use under the MIT License. However, if it helps you build profitable projects, we appreciate sponsorships to support continued development and maintenance.

Superpowers includes skills across multiple categories: Testing (test-driven-development), Debugging (systematic-debugging, verification-before-completion), Collaboration (brainstorming, writing-plans, subagent-driven-development, etc.), and Meta skills (writing-skills, using-superpowers). Each skill is designed to work across all supported agents.

The workflow follows a structured process: brainstorming → using-git-worktrees → writing-plans → subagent-driven-development → test-driven-development → requesting-code-review → finishing-a-development-branch. Each step is mandatory and ensures systematic development with proper testing and review cycles.

Subagent-driven development dispatches fresh subagents for each engineering task with a two-stage review process. First, the subagent checks for spec compliance, then for code quality. This enables autonomous work sessions where agents can work for hours without deviating from the plan.

Superpowers is built on four core principles: Test-Driven Development (write tests first), Systematic over ad-hoc (process over guessing), Complexity reduction (simplicity as primary goal), and Evidence over claims (verify before declaring success). This ensures reliable, maintainable code.

Superpowers enforces strict TDD with RED-GREEN-REFACTOR cycles, systematic debugging with root cause analysis, and comprehensive code review processes. It also emphasizes YAGNI (You Aren't Gonna Need It) and DRY (Don't Repeat Yourself) principles to eliminate unnecessary complexity.

While we don't generally accept contributions of new skills to maintain compatibility across all agents, you can create your own skills using the included `writing-skills` methodology. The framework provides the tools and best practices for developing new skills following our proven patterns.