PlugWiki

using-superpowers Skill for Claude: Always Check for Skills First

using-superpowers makes your agent check for a relevant skill before every response, even before clarifying questions.

Installs
276.8K
on skills.sh
Stars
310.6K
on GitHub
Files
7
besides SKILL.md
Updated
Oct 2026
last repo commit

using-superpowers is the skill meant for the start of any conversation. Its rule is to invoke any relevant or requested skill before any response or action, including clarifying questions, and to announce which skill it is using and why. Before plan mode it asks the agent to brainstorm first.

It puts process skills first: brainstorming for "let's build X" and systematic-debugging for "fix this bug". It points to a reference file for each supported agent and skips itself when dispatched as a subagent. It changes how your agent behaves from the first message, so expect more skill use.

Good for

  • Users of the Superpowers plugin who want its skills to trigger reliably
  • Developers who want brainstorming and debugging skills applied before work starts
  • Teams using Claude Code, Codex, Pi, Antigravity, Hermes Agent or Muse with the same skills

Install using-superpowers

Claude Code

  1. 1

    Copy this prompt.

    Install the using-superpowers skill in Claude Code. Run this command:
    
    /plugin install superpowers@claude-plugins-official
    
    Then check that it loaded and tell me how to use it.
  2. 2

    Paste it into Claude Code. It shows the change and asks before running it.

Install the plugin for the whole Superpowers set. This skill routes to brainstorming, systematic-debugging and other skills in the set.

From the project README. Source.

What's in the skill

When it loads

Use when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions

Sections in SKILL.md

  1. The Rule
  2. Skill Priority
  3. Red Flags
  4. Platform Adaptation
  5. User Instructions

Key facts

License
MIT
Installs on skills.sh
276,812 as of 2026-11-02
Part of plugin
superpowers
Last commit to the repo
2026-10-29

What it includes

  • Invoke relevant skills before any response

    Applies even if there is only a 1% chance a skill applies, and even before clarifying questions or exploring the code.

    Read · The Rule

  • Says "Using [skill] to [purpose]"

    Follows the skill exactly and creates a todo per checklist item.

    Try: “Let's build a CSV export feature.”

    Read · Announce and follow

  • Process skills before implementation skills

    Brainstorming first for building, systematic-debugging first for bugs, then domain skills.

    Read · Skill priority

  • Thoughts that mean stop and check for a skill

    A list of rationalizations such as "This is just a simple question" with the reply for each.

    Read · Red flags table

  • A reference file per agent

    SKILL.md points to reference files for Claude Code, Codex, Pi, Antigravity, Hermes Agent and Muse.

    Files · Platform adaptation

  • CLAUDE.md, AGENTS.md and your requests come first

    User instructions override skills, which override default behavior.

    Read · User instructions win

From the official docs

Questions

What does using-superpowers do?

It makes your agent look for a relevant skill before it responds to anything. It also sets the order: process skills such as brainstorming and systematic-debugging come before implementation skills.

How do I install it?

Use the install box at the top of this page, which has tabs for Claude Code, the Claude app and Codex. The install box uses the Superpowers plugin, which includes this skill.

Why does my agent announce "Using [skill]" all the time?

That is this skill. SKILL.md requires the agent to announce which skill it is using and why before it follows it.

Does it work in Codex and other agents?

SKILL.md lists reference files for Claude Code, Codex, Pi, Antigravity, Hermes Agent and Muse. The folder also carries a Gemini file that SKILL.md does not list.

Can I override it?

Yes. SKILL.md says your CLAUDE.md, AGENTS.md or direct requests take precedence over skills.

Is it safe?

Its folder holds only markdown files. The rule makes your agent invoke skills before every response, so it changes how the agent behaves in every conversation. See the audit results under key facts.

Information

updated Nov 2, 2026

Folder
skills/using-superpowers

obra/superpowersskills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 2
  1. obra/superpowers README · github.com
  2. using-superpowers SKILL.md · github.com

Updated Nov 2, 2026.