If a skill applies to your task, you do not have a choice. You must use it.
This is non-negotiable. This is not optional. You cannot rationalize your way out of this.
Instruction Priority
Superpowers skills override default system prompt behavior, butuser instructions always take highest priority：
- User's explicit instructions(CLAUDE.md, GEMINI.md, AGENTS.md, direct requests) - highest priority
- Superpowers skills - override default system behavior where they conflict
- Default system prompt - lowest priority
If CLAUDE.md, GEMINI.md, or AGENTS.md says "don't use TDD" and a skill says "always use TDD," follow the user's instructions. The user is in control.
How to Access Skills
In Claude Code: Use the Skill tool. When you invoke a skill, its content is loaded and presented to you - just follow it directly. Never use the Read tool to read skill files.
In Copilot CLI: Use the skill tool. Skills are auto-discovered from installed plugins.skill The Skill tool works the same way as Claude Code's
In Hermes Agent: Use the skill_view tool to load skills. Hermes supports three-level progressive loading:skills_list browse → skill_view(name) load full content → skill_view(name, path) view referenced files.
In Gemini CLI: Skills are activated via the activate_skill tool. Gemini loads skill metadata at session start and activates full content on demand.
In other environments: Check your platform documentation for how skills are loaded.
Platform Adaptation
Skills use Claude Code tool names. Non-CC platforms: see references/copilot-tools.md（Copilot CLI）、references/hermes-tools.md（Hermes Agent）、references/codex-tools.md（Codex）、references/qoder-tools.md(Qoder) for tool mapping. Gemini CLI users get the tool mapping automatically via GEMINI.md.
Rules
Invoke relevant or requested skills before any response or action. Even a 1% chance a skill might apply means you should invoke the skill to check. If it turns out the skill isn't right for the situation after invoking it, you don't need to use it.
digraph skill_flow {
"收到用户消息" [shape=doublecircle];
"即将进入 EnterPlanMode？" [shape=doublecircle];
"已经头脑风暴过？" [shape=diamond];
"调用头脑风暴技能" [shape=box];
"可能有技能适用？" [shape=diamond];
"调用 Skill 工具" [shape=box];
"宣布：'使用 [技能] 来 [目的]'" [shape=box];
"有检查清单？" [shape=diamond];
"为每个条目创建 TodoWrite 待办" [shape=box];
"严格遵循技能" [shape=box];
"响应（包括澄清）" [shape=doublecircle];
"即将进入 EnterPlanMode？" -> "已经头脑风暴过？";
"已经头脑风暴过？" -> "调用头脑风暴技能" [label="否"];
"已经头脑风暴过？" -> "可能有技能适用？" [label="是"];
"调用头脑风暴技能" -> "可能有技能适用？";
"收到用户消息" -> "可能有技能适用？";
"可能有技能适用？" -> "调用 Skill 工具" [label="是，哪怕只有 1%"];
"可能有技能适用？" -> "响应（包括澄清）" [label="确定不适用"];
"调用 Skill 工具" -> "宣布：'使用 [技能] 来 [目的]'";
"宣布：'使用 [技能] 来 [目的]'" -> "有检查清单？";
"有检查清单？" -> "为每个条目创建 TodoWrite 待办" [label="是"];
"有检查清单？" -> "严格遵循技能" [label="否"];
"为每个条目创建 TodoWrite 待办" -> "严格遵循技能";
}
Red Flags
These thoughts mean stop - you're rationalizing:
| Thought | Reality |
|---|---|
| "This is just a simple question" | Questions are tasks. Check for skills. |
| "I need to learn more context first" | Skill check comes before clarifying questions. |
| "Let me explore the codebase first" | Skills tell you how to explore. Check first. |
| "I can quickly check git/files" | Files lack conversation context. Check for skills. |
| "Let me gather information first" | Skills tell you how to gather information. |
| "This doesn't need a formal skill" | If a skill exists, use it. |
| "I remember this skill" | Skills are iteratively updated. Read the current version. |
| "This doesn't count as a task" | Action = task. Check the skill. |
| "The skill is making a big deal out of nothing" | Simple things become complicated. Use it. |
| "Let me just do this one thing first" | Check before doing anything. |
| "Doing it this way feels efficient" | Undisciplined action wastes time. Skills prevent this. |
| "I know what that means" | Knowing the concept ≠ using the skill. Invoke it. |
Skill Priority
When multiple skills may apply, use this order:
- Process skills take priority(brainstorming, debugging) - these determine how to handle the task
- Implementation skills come next(frontend design, mcp-builder) - these guide execution
"Let's build X" → brainstorm first, then use implementation skills. "Fix this bug" → debug first, then use domain-specific skills.
Chinese-Style Skill Routing
When the following scenarios are detected,you mustprioritize invoking the corresponding Chinese-style skills:
| Scenario | Skill to invoke |
|---|---|
| Code review and the team communicates in Chinese | superpowers:chinese-code-review |
| Using Gitee/Coding/Jihu GitLab | superpowers:chinese-git-workflow |
| Writing Chinese technical documentation or README | superpowers:chinese-documentation |
| Writing git commit messages (Chinese project) | superpowers:chinese-commit-conventions |
| Building an MCP server/tool | superpowers:mcp-builder |
Criteria for judgment:
- The project has Chinese comments, a Chinese README, or a .gitee directory → enable the Chinese skill series
- The commit history contains Chinese → use Chinese commit conventions
- The user communicates in Chinese → use Chinese for all output, and prioritize Chinese-style skills
Chinese-Style Skills and Translation Skillsare used in combination, not mutually exclusive. For example: when doing a code review, use requesting-code-review (process) + chinese-code-review (style) at the same time.
Skill Types
Rigid(TDD, debugging): follow strictly. Do not deviate from the discipline.
Flexible(patterns): adapt the principles according to context.
The skill itself will tell you which type it is.
User Instructions
Instructions say what to do, not how to do it. "Add X" or "fix Y" does not mean skipping the workflow.