研究检索只读github未标认证来源可访问许可证需确认审计通过

superpowers（superpowers 开发）

Agent Skill

用于辅助测试设计、自动化测试、用例整理和回归验证。它适合让 Agent 编写单元测试、端到端测试、测试计划或根据失败日志定位问题。使用时需要确认项目测试框架、运行命令和夹具数据，避免为了通过测试而改坏真实逻辑；涉及浏览器或外部服务时，应区分本地模拟、测试环境和生产环境。

总安装

1,223

周安装

GitHub Stars

公开资料未说明

下载量

428

CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息，不托管安装包，也不代用户安装。

GitHub

来源数

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手，先确认命令和权限，再让它执行。

请帮我安装这个 Agent Skill：superpowers（superpowers 开发）
来源仓库：https://github.com/hasmokan/the-reverse-turing-test
仓库路径：skills/superpowers
安装命令：
npx skills add https://github.com/hasmokan/the-reverse-turing-test --skill superpowers
安装前请先检查当前环境是否支持对应 CLI，并向我确认将要执行的命令、安装目录、联网范围和文件读写权限；确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill；本站只展示命令，不托管安装包，也不自动执行。

skills.shnpx skills

npx skills add https://github.com/hasmokan/the-reverse-turing-test --skill superpowers

简介

用于辅助设计自动化测试用例，覆盖单元、集成与端到端验证场景。

适合生成测试桩、模拟数据与回归测试计划，提升代码可靠性。
通过 GitHub 安装，需明确项目测试框架（如 Jest、Pytest）与运行命令。
测试执行可能依赖外部服务，应在隔离环境运行并设置超时控制。
superpowers 属于研究检索类 Skill，可作为该场景下的辅助能力补充。

SKILL.md

Superpowers - Agent Development Workflows

Comprehensive agent development skill collection optimized for AI-assisted coding workflows. Contains 14 skills across 5 categories, prioritized by workflow stage to guide development from ideation to completion.

When to Apply

Reference these skills when:

Starting a new feature or project (brainstorming, writing-plans)
Debugging any issue (systematic-debugging)
Writing or modifying code (test-driven-development)
Managing branches and isolation (using-git-worktrees)
Completing and integrating work (finishing-a-development-branch)
Receiving or requesting code review (receiving-code-review, requesting-code-review)

Skill Categories by Workflow Stage

Stage	Category	Skills	Priority
1	Ideation & Planning	`brainstorming`, `writing-plans`	CRITICAL
2	Implementation	`test-driven-development`, `subagent-driven-development`, `executing-plans`	CRITICAL
3	Debugging	`systematic-debugging`	HIGH
4	Code Review	`requesting-code-review`, `receiving-code-review`	HIGH
5	Completion	`verification-before-completion`, `finishing-a-development-branch`	HIGH
6	Utilities	`using-git-worktrees`, `dispatching-parallel-agents`, `using-superpowers`, `writing-skills`	MEDIUM

Quick Reference

1. Ideation & Planning (CRITICAL)

brainstorming - Collaborative design exploration before implementation
writing-plans - Create detailed implementation plans with bite-sized tasks

2. Implementation (CRITICAL)

test-driven-development - RED-GREEN-REFACTOR cycle for all code
subagent-driven-development - Execute plans with fresh subagent per task
executing-plans - Batch execution with checkpoints in separate session

3. Debugging (HIGH)

systematic-debugging - Four-phase root cause investigation before fixes

4. Code Review (HIGH)

requesting-code-review - Dispatch code reviewer subagent after implementation
receiving-code-review - Technical evaluation, not performative agreement

5. Completion (HIGH)

verification-before-completion - Evidence before claims, always
finishing-a-development-branch - Merge, PR, keep, or discard options

6. Utilities (MEDIUM)

using-git-worktrees - Isolated workspaces for parallel development
dispatching-parallel-agents - One agent per independent problem domain
using-superpowers - How to find and invoke skills
writing-skills - TDD applied to process documentation

Core Principles

Each skill enforces key discipline principles:

Principle	Skills	Iron Law
Test First	`test-driven-development`	No production code without failing test
Root Cause First	`systematic-debugging`	No fixes without investigation
Evidence First	`verification-before-completion`	No claims without verification
Plan First	`brainstorming`, `writing-plans`	No code without design

How to Use

Read individual skill files for detailed instructions:

brainstorming/SKILL.md
test-driven-development/SKILL.md
systematic-debugging/SKILL.md

Each skill file contains:

Overview and core principle
When to use (and when NOT to)
Step-by-step process
Common mistakes and red flags
Integration with other skills

Full Compiled Document

For the complete guide with all skills expanded: AGENTS.md

适合场景

用户想查找某类 Agent Skill 时

需要根据任务场景推荐可安装能力包时

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明，方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用；具体调用方式请以来源页面和 README 为准。

平台分布

Codex

35.5%

按下载量换算152

Claude

30.9%

按下载量换算132

Cursor

17.47%

按下载量换算75

Gemini CLI

9.97%

按下载量换算43

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容，本身偏只读；真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息，不托管安装包，不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源，正式发布前建议补源仓库或其他目录站核验。

来源信息

GitHub 仓库

github.com

/hasmokan/the-reverse-turing-test

SKILL.md 来源

github.com

/hasmokan/the-reverse-turing-test/tree/main/skills/superpowers

skills-sh

skills.sh

/hasmokan/the-reverse-turing-test/superpowers

继续浏览同类 Skills

更多研究检索 Skills（73,675）返回 Skills 目录

上一个 Skill

superpowers（superpowers 开发）

superpowers 用于查找、检索和筛选相关信息，适合在 Local Agent 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态，以及是否会触发联网、命令执行或文件读写。

下一个 Skill

superpowers-dispatching-parallel-agents（超级大国派遣平行特工）

superpowers-dispatching-parallel-agents 用于查找、检索和筛选相关信息，适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态，以及是否会触发联网、命令执行或文件读写。