obra
16 skills in 2 repos, most starred first.
superpowers Skill collection
An agentic skills framework & software development methodology that works.
314,664 stars on GitHub, Nov 9, 2026 · 15 skills · 1 bundle
-
Superpowers
Bundle of 15 skills- Use it when:
- You want the agent to follow a full development process instead of jumping straight to code.
- It does:
- Chains its skills into one workflow: brainstorm, plan, build test-first on its own branch, review, then merge.
using-superpowers, brainstorming, using-git-worktrees, writing-plans, subagent-driven-development, executing-plans, test-driven-development, requesting-code-review, receiving-code-review, finishing-a-development-branch, systematic-debugging, verification-before-completion, dispatching-parallel-agents, writing-skills, diagnosing-superpowers
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 422K
- #50
- Weekly Downloads
- 8.9K
- #65
-
brainstorming
Part of Superpowers- Use it when:
- You're about to build or plan something new, in software or not, and haven't spelled out what you want.
- It does:
- Asks one open question per message and plays back what it understood, part by part, until the picture is complete.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 422K
- #50
- Weekly Downloads
- 8.2K
- #67
-
systematic-debugging
Part of Superpowers- Use it when:
- A bug, failing test or odd behavior shows up, especially when a quick fix is tempting.
- It does:
- Forbids any fix until the root cause is found, through four phases in order.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 325K
- #65
- Weekly Downloads
- 8.9K
- #65
-
writing-plans
Part of Superpowers- Use it when:
- You have a spec or approved design for a multi-step change, before any code is written.
- It does:
- Breaks the work into small test-first tasks, each naming its files, function signatures, tests and the commands that prove it.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 298K
- #66
- Weekly Downloads
- 7K
- #89
-
using-superpowers
Only works as part of Superpowers- Use it when:
- You install the Superpowers bundle: it's the bundle's entry point.
- It does:
- Makes the agent check for a matching skill before every reply, even before a clarifying question.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 283K
- #69
- Weekly Downloads
- 6K
- top 10%
-
requesting-code-review
Part of Superpowers- Use it when:
- A task or feature is done, before merging, or when you're stuck and want fresh eyes.
- It does:
- Sends a reviewer subagent the commit range and the plan, with none of the session's history.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 276K
- #70
- Weekly Downloads
- 6.5K
- top 9%
-
test-driven-development
Part of Superpowers- Use it when:
- The agent implements any feature, bug fix, refactor or behavior change.
- It does:
- Makes it write one failing test, watch it fail, write the least code that passes, then clean up.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 274K
- #71
- Weekly Downloads
- 6.9K
- #90
-
- Use it when:
- The agent is about to say work is done, fixed or passing, or to commit, open a PR or move on.
- It does:
- Makes it run the command that proves the claim, read the full output, and only then report what it shows.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 269K
- #73
- Weekly Downloads
- 8.1K
- #69
-
executing-plans
Only works as part of Superpowers- Use it when:
- You have a plan from writing-plans and want the cheaper way to run it, or the agent can't start subagents.
- It does:
- Works through every task itself, test-first, on its own branch, without stopping to check in between tasks.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 257K
- #75
- Weekly Downloads
- 6.3K
- top 9%
-
subagent-driven-development
Only works as part of Superpowers- Use it when:
- You have a plan with mostly independent tasks and want a review gate on every task.
- It does:
- Hands each task to a fresh implementer subagent, then has a reviewer subagent check it against the spec and for quality.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 251K
- #77
- Weekly Downloads
- 6.2K
- top 9%
-
receiving-code-review
Part of Superpowers- Use it when:
- You get review feedback, from a person or a reviewer agent, especially if it seems unclear or wrong.
- It does:
- Makes the agent check each suggestion against the code before changing anything, and push back with reasons when it's wrong.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 234K
- #81
- Weekly Downloads
- 6.1K
- top 9%
-
- Use it when:
- You have two or more problems that don't share state, such as test failures in unrelated files.
- It does:
- Gives each problem its own subagent with a focused, self-contained prompt, and starts them all at once.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 231K
- #82
- Weekly Downloads
- 5.9K
- top 10%
-
writing-skills
Part of Superpowers- Use it when:
- You're writing or editing a skill and want proof that agents actually follow it.
- It does:
- Tests a skill like code: run a subagent without it and watch it go wrong, then write the skill and check the subagent now behaves.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 231K
- #83
- Weekly Downloads
- 5.8K
- top 10%
-
using-git-worktrees
Part of Superpowers- Use it when:
- You're starting feature work or about to run a plan, and want your current branch left untouched.
- It does:
- Sets up a git worktree, a second checkout of the repo on its own branch, after asking you.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 230K
- #85
- Weekly Downloads
- 5.9K
- top 10%
-
- Use it when:
- The work on a branch is done and you need to decide how to integrate it.
- It does:
- Runs the full test suite, then offers three choices: merge locally, push and open a PR, or keep the branch.
- Age
- 9 months
- Stars
- 315K
- #4
- Total Downloads
- 227K
- #86
- Weekly Downloads
- 5.9K
- top 10%
-
diagnosing-superpowers
Only works as part of Superpowers- Use it when:
- A Superpowers session repeated work, ignored its plan, cost too much or never fired a skill, and you want to know why.
- It does:
- Asks what went wrong, finds the session's transcripts, and runs seven analyst subagents in parallel, one per angle such as plan adherence or cost.
- Age
- 7 weeks
- Stars
- 315K
- #4
- Total Downloads
- 30.4K
- top 22%
- Weekly Downloads
- 5K
- top 10%
episodic-memory Project
Semantic search for Claude Code, Codex, Cursor, opencode, and Oh My Pi (OMP) conversations. Remember past discussions, decisions, and patterns.
499 stars on GitHub, Nov 9, 2026 · 1 skill
-
- Use it when:
- A task resembles past work, or you say "last time", "we discussed" or "why did we", and the answer lives in an earlier conversation.
- It does:
- In Claude Code, hands the search to episodic-memory's search-conversations subagent; elsewhere it calls the search and read tools itself.
- Age
- 9 months
- Stars
- 499
- bottom 13%
- Total Downloads
- 9.2K
- bottom 47%
- Weekly Downloads
- 16
- bottom 7%