obra/superpowers

Source
GitHub
First trending
Category
Skills
GitHub stars
292,514
Main language
Shell

This page introduces an external open-source repository. It is not an HDATF product.

obra/superpowers

What it does

Superpowers is a software development method for coding agents, built from composable skills. The agent first draws out a spec, gets the design and an implementation plan approved, then runs subagent-driven development with red/green TDD.

How it helps ATF

Its order of spec, approved plan, then implementation by subagents is worth comparing with how Harness keeps a goal and checks results with evidence. The skill set is a reference for packaging repeatable work steps.

License

MIT Permissive. Commercial use and changes are allowed if the copyright notice is kept.

More in this category

  • ComposioHQ/awesome-claude-skills
    A curated list of Claude skills and plugins grouped into areas such as document processing, development tools, data analysis, business and communication, plus a plugin that connects Claude to outside apps through Composio.
  • bmad-code-org/BMAD-METHOD
    A method for AI-driven software development, delivered as skills for AI coding tools such as Claude Code and Codex. It takes an idea or change request through briefs, specifications and architecture to working software, keeping decisions explicit.
  • kepano/obsidian-skills
    Agent skills that teach agents such as Claude Code, Codex and OpenCode to work with Obsidian through its CLI and open formats, including Markdown, Bases and JSON Canvas. They follow the Agent Skills specification.
  • K-Dense-AI/scientific-agent-skills
    A library of agent skills for scientific work in biology, chemistry, medicine and drug discovery, with access to scientific databases. It follows the open Agent Skills standard, so agents such as Claude Code, Codex and Cursor can use it.
  • OthmanAdi/planning-with-files
    An agent skill that keeps task_plan.md, findings.md and progress.md on disk and uses lifecycle hooks to re-inject the plan every turn, so coding agents can resume after /clear, crashes or context compaction.

Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.

View on GitHub (opens in a new tab)