superpowers-zh plugin
AI 编程超能力中文增强版:21 个 skills(15 翻译 + 4 中国原创 + 2 上游历史保留),支持 Claude Code / Hermes Agent / Cursor / Claw Code / Qoder 等 26 款工具
How to install the superpowers-zh plugin
Run both commands inside Claude Code. The first adds the marketplace once; the second installs the plugin.
/plugin marketplace add jnMetaCode/superpowers-zh
/plugin install superpowers-zh@superpowers-zhWhat's inside superpowers-zh
22 parts read from the plugin's own files.
Skills 21
Claude loads it when the task matches- brainstorming
Helps an agent clarify user intent and requirements through open-ended questioning and visual prototyping before starting any creative or technical work.
- chinese-code-review
Provides templates, priority labeling, and communication strategies for conducting effective code reviews in Chinese-speaking development teams.
- chinese-commit-conventions
Provides configuration references for Chinese-language Conventional Commits, including commitlint, husky, commitizen, and changelog setups.
- chinese-documentation
Provides comprehensive guidelines for Chinese technical writing, including spacing, punctuation, terminology handling, and API documentation templates.
- chinese-git-workflow
Provides configuration guides and workflow standards for domestic Git platforms including Gitee, Coding.net, JiHu GitLab, and CNB.
- diagnosing-superpowers
Analyzes session transcripts to provide evidence-based reports on performance issues, plan adherence, and cost inefficiencies.
- dispatching-parallel-agents
Delegates independent tasks to specialized agents with isolated contexts to resolve multiple issues concurrently.
- executing-plans
Enables an agent to execute a pre-defined implementation plan task-by-task using inline execution, TDD, and a ledger for continuity.
- finishing-a-development-branch
Provides a structured workflow for verifying tests, detecting environments, and executing integration options like local merging or creating pull requests.
- mcp-builder
Provides a systematic methodology for designing, implementing, testing, and deploying Model Context Protocol servers and tools.
- receiving-code-review
Provides a structured framework for evaluating, verifying, and responding to code review feedback with technical rigor instead of social politeness.
- requesting-code-review
Dispatches a specialized sub-agent to perform structured code reviews by providing specific git SHAs and task descriptions instead of full session history.
- subagent-driven-development
Executes implementation plans by assigning isolated subagents to individual tasks with mandatory per-task and final branch reviews.
- systematic-debugging
Executes a four-stage investigative process to identify root causes and implement verified fixes for technical bugs and system anomalies.
- test-driven-development
Follows the Red-Green-Refactor cycle to write failing tests before implementing the minimum amount of code required to pass.
- using-git-worktrees
Creates isolated development environments using native tools or git worktree commands to prevent interference with the current working directory.
- using-superpowers
Enforces a strict protocol for agents to search for and invoke specific skills before performing any task, clarification, or exploration.
- verification-before-completion
Enforces a strict verification protocol requiring empirical evidence from command outputs before declaring any task, fix, or test as complete.
- workflow-runner
Executes multi-role YAML workflows within AI coding tools by using the current session's LLM to play different agent roles sequentially.
- writing-plans
Creates detailed implementation plans for multi-step software tasks by breaking specifications into small, testable, and unambiguous technical steps.
- writing-skills
Apply test-driven development to create, edit, and validate technical reference guides for AI agents.
Hooks 1
Runs on these events- hooks
No description in the file.
Using superpowers-zh in Claude Code
Skills need no command. Claude reads them when your request matches what they describe.
Update, turn off or remove
- Update the plugin (shell)
- claude plugin update superpowers-zh@superpowers-zh
- Refresh the marketplace
- /plugin marketplace update superpowers-zh
- Turn it off for now
- /plugin disable superpowers-zh@superpowers-zh
- Remove it
- /plugin uninstall superpowers-zh@superpowers-zh
Skills in superpowers-zh
- chinese-code-reviewjnMetaCode/superpowers-zh
Provides templates, priority labeling, and communication strategies for conducting effective code reviews in Chinese-speaking development teams.
8.5Kstars - chinese-commit-conventionsjnMetaCode/superpowers-zh
Provides configuration references for Chinese-language Conventional Commits, including commitlint, husky, commitizen, and changelog setups.
8.5Kstars - chinese-documentationjnMetaCode/superpowers-zh
Provides comprehensive guidelines for Chinese technical writing, including spacing, punctuation, terminology handling, and API documentation templates.
8.5Kstars - chinese-git-workflowjnMetaCode/superpowers-zh
Provides configuration guides and workflow standards for domestic Git platforms including Gitee, Coding.net, JiHu GitLab, and CNB.
8.5Kstars - executing-plansjnMetaCode/superpowers-zh
Enables an agent to execute a pre-defined implementation plan task-by-task using inline execution, TDD, and a ledger for continuity.
3 files8.5Kstars - mcp-builderjnMetaCode/superpowers-zh
Provides a systematic methodology for designing, implementing, testing, and deploying Model Context Protocol servers and tools.
8.5Kstars - subagent-driven-developmentjnMetaCode/superpowers-zh
Executes implementation plans by assigning isolated subagents to individual tasks with mandatory per-task and final branch reviews.
7 files8.5Kstars - systematic-debuggingjnMetaCode/superpowers-zh
Executes a four-stage investigative process to identify root causes and implement verified fixes for technical bugs and system anomalies.
11 files8.5Kstars - test-driven-developmentjnMetaCode/superpowers-zh
Follows the Red-Green-Refactor cycle to write failing tests before implementing the minimum amount of code required to pass.
2 files8.5Kstars - using-git-worktreesjnMetaCode/superpowers-zh
Creates isolated development environments using native tools or git worktree commands to prevent interference with the current working directory.
8.5Kstars - workflow-runnerjnMetaCode/superpowers-zh
Executes multi-role YAML workflows within AI coding tools by using the current session's LLM to play different agent roles sequentially.
8.5Kstars - writing-skillsjnMetaCode/superpowers-zh
Apply test-driven development to create, edit, and validate technical reference guides for AI agents.
7 files8.5Kstars - brainstormingjnMetaCode/superpowers-zh
Helps an agent clarify user intent and requirements through open-ended questioning and visual prototyping before starting any creative or technical work.
8 files8.5Kstars - diagnosing-superpowersjnMetaCode/superpowers-zh
Analyzes session transcripts to provide evidence-based reports on performance issues, plan adherence, and cost inefficiencies.
16 files8.5Kstars - dispatching-parallel-agentsjnMetaCode/superpowers-zh
Delegates independent tasks to specialized agents with isolated contexts to resolve multiple issues concurrently.
8.5Kstars - finishing-a-development-branchjnMetaCode/superpowers-zh
Provides a structured workflow for verifying tests, detecting environments, and executing integration options like local merging or creating pull requests.
8.5Kstars - receiving-code-reviewjnMetaCode/superpowers-zh
Provides a structured framework for evaluating, verifying, and responding to code review feedback with technical rigor instead of social politeness.
8.5Kstars - requesting-code-reviewjnMetaCode/superpowers-zh
Dispatches a specialized sub-agent to perform structured code reviews by providing specific git SHAs and task descriptions instead of full session history.
2 files8.5Kstars - using-superpowersjnMetaCode/superpowers-zh
Enforces a strict protocol for agents to search for and invoke specific skills before performing any task, clarification, or exploration.
10 files8.5Kstars - verification-before-completionjnMetaCode/superpowers-zh
Enforces a strict verification protocol requiring empirical evidence from command outputs before declaring any task, fix, or test as complete.
8.5Kstars - writing-plansjnMetaCode/superpowers-zh
Creates detailed implementation plans for multi-step software tasks by breaking specifications into small, testable, and unambiguous technical steps.
8.5Kstars
superpowers-zh plugin questions
How do I install the superpowers-zh plugin?
In Claude Code run /plugin marketplace add jnMetaCode/superpowers-zh, then /plugin install superpowers-zh@superpowers-zh.
What does the superpowers-zh plugin include?
superpowers-zh bundles 21 skills, hooks.
How do I update or remove the superpowers-zh plugin?
Update it with claude plugin update superpowers-zh@superpowers-zh in your shell, or from the Installed tab of /plugin. Remove it with /plugin uninstall superpowers-zh@superpowers-zh.