using-superpowers

Enforces skill discovery and invocation before any AI response or action.

Updated Apr 7, 2026
One-click install
npx skills add https://github.com/NT-boop-star/BRMV-tract --skill using-superpowers-nt-boop-star
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: using-superpowers
Source: https://github.com/NT-boop-star/BRMV-tract/tree/main/antigravity/skills/using-superpowers
Command: npx skills add https://github.com/NT-boop-star/BRMV-tract --skill using-superpowers-nt-boop-star

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI agents often skip available skills and respond directly, missing established workflows and producing inconsistent results. This Skill establishes a mandatory protocol that requires checking for and invoking relevant skills before any response, including clarifying questions. ## Core Features & Use Cases - Mandatory Skill Invocation: Requires invoking any potentially applicable skill (even at 1% relevance) before responding or acting. - Cross-Platform Tool Mapping: Provides reference tables mapping Claude Code tools to Copilot CLI, Codex, and Gemini CLI equivalents. - Priority and Red-Flag Rules: Defines instruction priority (user instructions over skills over system defaults) and lists rationalization patterns to avoid. - Use Case: When a user asks an agent to fix a bug, the agent first invokes the debugging skill, announces its use, creates todos from the skill checklist, and follows the workflow exactly. ## Quick Start Ask the agent to start any task and it will check for applicable skills, invoke them, and announce which skill it is using before responding.

Frequently Asked Questions about using-superpowers

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make an AI agent always use available skills before responding?

Install this skill so it loads at conversation start. It instructs the agent to invoke any potentially relevant skill via the Skill tool before any response, including clarifying questions, and to announce which skill it is using.

How do Claude Code skills work in Copilot CLI or Gemini CLI?

Copilot CLI uses the skill tool with auto-discovered plugins, and Gemini CLI uses the activate_skill tool. The references directory provides full tool-mapping tables translating Claude Code tool names to each platform's equivalents.

Do user instructions override skill workflows like TDD?

Yes. User instructions in CLAUDE.md, GEMINI.md, AGENTS.md, or direct requests take highest priority. Skills override default system behavior, but if a user says not to use TDD, the agent follows the user.

What happens when multiple skills could apply to one task?

Process skills like brainstorming or debugging run first because they determine how to approach the task. Implementation skills such as frontend-design run second to guide execution.

Can subagent-based skills run in Gemini CLI or Codex?

Gemini CLI has no subagent support, so those skills fall back to single-session execution. Codex supports subagents through spawn_agent when multi_agent is enabled in config.toml, using worker agents with filled prompt templates.