superpowers

Routes coding agents through structured planning, TDD, debugging, review, and verification workflows.

1|Updated Sep 10, 2026
One-click install
npx skills add https://github.com/thienty1207/Hotel_Staff --skill superpowers-thienty1207
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: superpowers
Source: https://github.com/thienty1207/Hotel_Staff/tree/main/.baron/core/skills/superpowers
Command: npx skills add https://github.com/thienty1207/Hotel_Staff --skill superpowers-thienty1207

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? AI coding agents often jump straight into code without a plan, skip tests, guess at bug fixes, and claim completion without verification. This Skill enforces a disciplined software development methodology by routing every task to the correct workflow sub-skill before any action is taken. ## Core Features & Use Cases - Workflow Routing: A routing index maps task shapes (feature work, bug fixes, plan execution, code review) to exactly one of 14 vendored sub-skills such as test-driven-development, systematic-debugging, writing-plans, and subagent-driven-development. - Process Discipline Contracts: Hard rules protect workflow boundaries, including plan-scoped ledger isolation, bounded review-fix rounds (rounds 1-5 with escalation), and behavior-protecting test requirements. - Vendored and Pinned: Sub-skills are pinned to upstream Superpowers v6.2.0 with recorded tree hashes, a local-only patch removing telemetry, and no runtime downloads. - Use Case: When asked to fix a failing test, the agent loads systematic-debugging, completes root-cause investigation before proposing fixes, writes a failing regression test via TDD, and verifies the fix before claiming completion. ## Quick Start Ask the agent to plan and implement a new feature, and it will load the brainstorming and writing-plans sub-skills before writing any code.

Frequently Asked Questions about superpowers

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

FAQPage Schema
How do I make an AI coding agent follow TDD?

Load the test-driven-development sub-skill, which enforces the red-green-refactor cycle: write a failing test, watch it fail, write minimal code to pass, then refactor. Code written before a failing test must be deleted and rewritten from tests.

What is the Superpowers workflow for coding agents?

Superpowers is a methodology that routes tasks through brainstorming, writing-plans, subagent-driven-development or executing-plans, test-driven-development, code review, and finishing-a-development-branch. The agent checks for a relevant skill before any response or action.

How does systematic debugging differ from quick fixes?

Systematic debugging requires completing root cause investigation before proposing any fix, forming a single hypothesis, and testing it minimally. After three failed fixes, it mandates questioning the architecture instead of attempting a fourth fix.

Which coding agents support Superpowers skills?

The upstream documentation lists Claude Code, Codex CLI and App, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, Factory Droid, Kimi Code, Antigravity, and Pi. In this repository it is vendored under Baron as the sole workflow core.

When should I not dispatch parallel subagents?

Avoid parallel dispatch when failures are related, when understanding requires full system context, during exploratory debugging, or when agents would share state such as editing the same files. Use one agent per independent problem domain only.