Superpowers

A software development methodology for coding agents, packaged as composable skills that make the agent clarify a spec, plan, and build with test-driven development and subagents.

Agent skills

Visit website github.com

Superpowers is a set of skills and starting instructions for AI coding agents. When the agent sees you are building something, it first asks what you are trying to do, presents a spec in short sections for approval, then writes an implementation plan.

Once you approve, it runs a subagent-driven development process in which agents work through each task while their work is inspected and reviewed. The plans stress red/green TDD, YAGNI and DRY, and the skills trigger automatically.

It installs as a plugin for Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode and other agent harnesses. It is MIT licensed.

More in Agent skills

See all 581

Provenance gate
github.com
A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.
- Agent skills
- AI evals
Expected Harm Calculus (EHC-4): Asimov's Four Laws for Autonomous AI Agents
github.com
A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.
- Agent skills
Vitruvius
github.com
An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.
- Agent skills
- AI evals

More in Agent skills

Provenance gate

Expected Harm Calculus (EHC-4): Asimov's Four Laws for Autonomous AI Agents

Vitruvius

Cookies