logoSeailife
  • Submit Project
  • Pricing
  • Blog
Sign inSign up
Sign in
logoSeailife

© 2026 Seailife. All rights reserved.

Built with Open Launch - The first complete open source alternative to Product Hunt.

For AI agents: llms.txt

Powered by Open-LaunchPowered by Open-Launch

Discover

  • Trending
  • Categories
  • Submit Project

Resources

  • Pricing
  • Sponsors
  • Blog

Legal

  • Terms of Service
  • Privacy Policy
  • Submission Guidelines

Connect

  • [email protected]
Superpowers Logo

Superpowers

Developer ToolsOpen SourceAI Agents
Visit
Superpowers - Product Image

Superpowers is a free MIT skills framework that makes coding agents brainstorm, plan, test first and review as they build, in Claude Code, Codex and Cursor.

What it is: a fixed way of working, not a single tool

The repo is by Jesse Vincent and the Prime Radiant team, MIT licensed. The README calls it a complete software development methodology for coding agents, built on composable skills plus starting instructions that make the agent use them. The skills fire on their own when the agent sees what you are doing, so you do not have to invoke anything.

  • It does not jump into code. It first asks what you are really trying to do, then shows you the spec in chunks short enough to read. Small changes are agreed in chat, and a project gets a written design document.
  • After you approve the design, it writes a plan of tasks of 2 to 5 minutes each, with exact file paths, complete code and verification steps, stressing red/green TDD, YAGNI and DRY.
  • When you say "go", it either sends a fresh subagent per task with a review after each, which the README calls the most thorough, or runs the tasks inline with one review of the whole branch at the end, which it calls the cheapest. The README says it is not uncommon for the agent to work on its own for a couple of hours without leaving the plan. That is the authors' description.

The seven-step workflow and the skills behind it

  • The README's order is brainstorming, using-git-worktrees, writing-plans, subagent-driven-development or executing-plans, test-driven-development, requesting-code-review and finishing-a-development-branch. It says these are mandatory workflows, not suggestions.
  • Test-driven development enforces red, green, refactor: write a failing test, watch it fail, write the minimum code, watch it pass, commit. The README says code written before the tests is deleted.
  • systematic-debugging is a four-phase root-cause process. verification-before-completion makes the agent check that something is actually fixed. Anthropic's marketplace page adds that it triggers an architecture review after three failed fix attempts.
  • The skills/ folder holds 15 skills, including dispatching-parallel-agents, receiving-code-review, writing-skills for making new ones and diagnosing-superpowers, which reads a session transcript and reports what went wrong with line-level evidence.
  • The README's philosophy has four lines: tests first, process over guessing, simplicity as the main goal, and evidence over claims.

Where you can install it

  • Claude Code:/plugin install superpowers@claude-plugins-official from Anthropic's official marketplace, or add obra/superpowers-marketplace and install superpowers@superpowers-marketplace.
  • Others the README gives steps for: Codex app and CLI through the official Codex plugin marketplace, Cursor with /add-plugin superpowers, Gemini CLI, GitHub Copilot CLI, Qwen Code, Kimi Code, Grok Build CLI, Factory Droid, Devin CLI, OpenCode, Pi, Antigravity, Muse and Hermes Agent with hermes plugins install obra/superpowers --enable.
  • The README says to install it separately for each agent you use. Updates depend on the agent and are often automatic. To check an install worked, it suggests sending "Let's make a react todo list" and seeing whether brainstorming starts before any code is written.

Limits and things to know

  • The maintainers say they do not generally accept new skills, that any change to a skill must work across all the supported agents, and that they do not accept agent-written pull requests for issues someone else filed unless a maintainer asked for one.
  • There is optional telemetry: the Prime Radiant logo in brainstorming's visual companion is loaded from their website and includes the Superpowers version, nothing about your project, prompt or agent. Set SUPERPOWERS_DISABLE_TELEMETRY to turn it off. It also honors Claude Code's DISABLE_TELEMETRY and CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC opt-outs.
  • The old superpowers-skills repo, community-editable skills for the plugin, is archived.

How old it is

  • This is not a new project. The repository was created on October 9, 2025, announced on the author's blog that day, and its latest release on GitHub is v7.0.2 from November 3, 2026.
  • On November 7 it had about 313,000 stars, about 27,900 forks and 342 open issues and pull requests together.
  • Seailife did not install it or run any session with it. The workflow and install steps above are from the README and the Claude marketplace page.

Comments

Submitted by

Seailife

Seailife

Launch Date
2026-11-11
Platform
other
Pricing
free
Socials

Tech Stack

#Shell

Sponsors

Become a Sponsor

Get your brand featured here

Similar tools

  • DeepSeek Super-Agent logo
    DeepSeek Super-Agentfree

    DeepSeek Super-Agent is an open-source research prototype of a coding agent that watches one GitHub repo and opens pull requests when allowed people ask.

  • pixelfx logo
    pixelfxfree

    pixelfx is a free MIT library of 244 pixel art game effects drawn by code, with sprite sheet export and a Claude Code skill for making new ones.

  • Ponytail logo
    Ponytailfree

    Ponytail is an open-source skill that makes Claude Code and other coding agents write the smallest complete change, for developers tired of bloated diffs.

  • Day Trader Skill logo
    Day Trader Skillfree

    Day Trader Skill is a free set of 11 Claude skills that scan charts, plan entries and exits and size risk for day trades, with paper trading by default.

  • Instagram Agent Skill logo
    Instagram Agent Skillfree

    Instagram Agent Skill is a free set of 13 Claude skills that write Reels, captions and weekly plans for an Instagram account, for creators who post themselves.

More in AI Agents