superpowers

Load and invoke skills before every assistant response.

Updated Aug 27, 2026

Search Tags:#guidance #safety #priority #workflow-control #skill-invocation #tool-usage

Ready to run

npx skills add https://github.com/mengdehong/opencode-dotfiles --skill superpowers-mengdehong

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: superpowers
Source: https://github.com/mengdehong/opencode-dotfiles/tree/main/skills/superpowers
Command: npx skills add https://github.com/mengdehong/opencode-dotfiles --skill superpowers-mengdehong

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Ensures that Skills are discovered and loaded before any reply, enforcing a reliable and safe starting state for all conversations.

Core Features & Use Cases

Enforces pre-response Skill invocation to determine applicable capabilities and safety constraints.
Supports consistent prioritization where user instructions override superpowers rules, guiding interaction flow.
Facilitates quick onboarding of new Skills by always loading the Skill registry at session start.

Quick Start

Invoke the Skill tool at the start of every conversation to load and apply the relevant skills before any response.

Frequently Asked Questions about superpowers

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

FAQPage Schema

How do I ensure skills are loaded before an assistant response is generated?▼

To ensure skills are loaded before an assistant response, invoke the Skill tool at the start of every conversation to enforce a reliable and safe starting state. This pre-response invocation determines applicable capabilities and safety constraints.

What is skill-driven planning and how does it work?▼

Skill-driven planning enables consistent prioritization by loading the Skill registry at session start before any reply. It works by honoring the Skill tool interface with explicit priority rules where user instructions override rules to guide interaction flow.

Can I use explicit priority rules to override skill invocation behavior?▼

Yes, you can use explicit priority rules to override skill invocation behavior. Consistent prioritization is supported where user instructions override rules, guiding the interaction flow and determining how and when skills are invoked.

How do I onboard new skills at the start of a session?▼

To onboard new skills at the start of a session, invoke the Skill tool at the beginning of every conversation. This facilitates quick onboarding by always loading the Skill registry before any response is generated.

Why does the assistant need to invoke skill tools before replying?▼

The assistant must invoke skill tools before replying to enforce pre-response skill invocation and determine applicable capabilities and safety constraints. This solves the problem of ensuring skills are discovered and loaded for a reliable starting state.

What are the limitations of enforcing pre-response skill invocation?▼

A limitation of enforcing pre-response skill invocation is that it requires the Skill tool interface to be strictly honored. It applies at the start of every conversation and for clarifications, necessitating skill tool usage before any execution.