Guide · Agent skill
How to use the Using Superpowers agent skill (using-superpowers)
Updated · written with AI from the project's own documentation and checked against our catalogue data
The using-superpowers skill is a meta-skill from the obra/superpowers project under the MIT licence. It sits at the start of a conversation and tells your AI agent to look for other skills that might apply to a request and use them before it answers, asks clarifying questions, or takes action. This guide explains what it changes, how to install it, how to trigger it, and where it can get in the way, so you can decide whether it fits your setup.
Using Superpowersobra/superpowers · MIT Download the skill Quick start
- 1Download the ZIP. Download the using-superpowers ZIP from this page.
- 2Unzip into the skills folder. Unzip it into ~/.claude/skills/ for personal use, or into the project's .claude/skills/ folder for one project.
- 3Or upload in the Claude apps. In the Claude apps, open Settings, go to Capabilities, and upload the skill under Skills.
- 4Install other skills you use. Make sure the skills you want the agent to apply are also installed, since this skill has little effect on its own.
- 5Start a conversation and make a request. Ask for a task in plain language, such as building a feature or fixing a bug, and watch whether the agent names the skill it is using.
- 6Check the order of skills. Confirm that process skills like brainstorming or systematic-debugging run before implementation skills.
- 7Adjust project instructions if needed. If the agent over-applies the skill, state in your project instruction file when quick answers should skip skill checks.
What the using-superpowers skill makes the agent do
Agent skills are instruction packages that an agent can load for specific kinds of work. using-superpowers does not add a new capability by itself. Instead, it changes how the agent decides when to use the capabilities it already has.
The skill tells the agent to invoke relevant or requested skills before any response or action, including clarifying questions, exploring the codebase, or checking files. It also sets an order when several skills apply: process skills such as brainstorming or systematic-debugging come first, and implementation skills such as frontend-design carry out the chosen approach.
- Checks for applicable skills before answering or acting, rather than after.
- Orders multiple skills so process skills run before implementation skills.
- Includes a table of rationalizations, such as "This is just a simple question", that the agent is told to resist.
- Points the agent to platform-specific reference notes for Claude Code, Codex, Pi, Antigravity, Hermes Agent and Muse.
Install the using-superpowers skill
Download the ZIP from this page. You can install it in one of two ways. For a personal setup, unzip it into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder. The skill folder should contain SKILL.md and the references/ folder.
If you use the Claude apps instead, open Settings, go to Capabilities, and upload the skill under Skills.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload
Trigger the skill with natural requests
You do not need a special command. The skill's description tells the agent to use it when starting any conversation, so it is loaded at the start of a session. You can also ask for it directly in plain language, for example by naming the workflow you want.
The source names two example workflows: "Let's build X" makes the agent run brainstorming first, then implementation skills, and "Fix this bug" makes it run systematic-debugging first, then domain skills. Before entering plan mode, the agent is told to brainstorm first if it has not already.
- "Let's build a CSV export for orders" (brainstorming first, then implementation skills)
- "Fix this bug where the login form submits twice" (systematic-debugging first, then domain skills)
- Use the skill's name in a request if you want a specific skill applied
Know what it needs and what it runs
The skill lists no requirements. It does not run scripts and does not use the network. Its effect depends on other skills being installed: with no other skills available, it has little effect on its own.
Because the skill has no scripts or network access, the main safety consideration is behavioural rather than technical. Its wording is very forceful, so an agent may over-apply it. Explicit user or project instructions, such as CLAUDE.md or AGENTS.md, take precedence over skills according to the file itself.
- Requirements: none listed
- Runs scripts: no
- Uses network: no
- Check your project instructions if they conflict with the skill's rules
Tune the strictness and avoid over-application
The skill pushes strongly toward checking for skills before every response. That is useful when an agent keeps skipping relevant skills, but it can add overhead to simple questions. If the agent becomes slow or rigid, the source says user and project instructions override skills, so put any exceptions in your project instruction file.
The file also says that if a skill turns out to be wrong for the situation, you do not have to use it. Review the agent's first few responses in a new project to see how often it checks skills for trivial requests.
- Keep project instructions explicit about when you want quick answers without skill checks
- Use it when you already have a set of skills and want them applied consistently
- Test on a small task before using it across a large codebase
What people use it for
- Keeping a brainstorm-plan-implement-debug workflow consistent across coding sessions
- Making an agent run systematic-debugging before it edits code when a bug is reported
- Stopping an agent from answering a request before checking whether a relevant skill exists
- Applying a team's own set of skills consistently across several projects
Frequently asked questions
Is the using-superpowers skill free?
The skill is released under the MIT licence, as stated on this page. Check the licence file in the source repository for the full terms.
What does using-superpowers require?
The skill lists no requirements. It does not run scripts and does not use the network, but it works best when other skills are also installed.
How do I install using-superpowers in Claude Code?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into your project's .claude/skills/ folder. The folder should contain SKILL.md and references/.
How do I install it in the Claude apps?
Open Settings, go to Capabilities, and upload the skill under Skills. The skill files come from the ZIP downloaded from this page.
Why does my agent check skills even for simple questions?
The skill tells the agent to check for relevant skills before any response, including simple questions. If that adds too much overhead, state in your project instructions when quick answers should skip skill checks, since user and project instructions take precedence over skills.
Does using-superpowers add new abilities to my agent?
No. It changes when and how the agent uses the skills it already has. If no other skills are installed, it has little effect on its own.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use using-superpowers if you already have agent skills and want your agent to apply them consistently, especially for a structured brainstorm, plan, implement and debug workflow. Pick something else if you want quick, flexible answers to simple questions, or if you have no other skills installed, since the skill's strong wording can add overhead and rigidity.
Download the skill