Use when starting any conversation — establishes how to find and use skills, requiring skill loading before any response including clarifying questions
If there is even a 1% chance a skill might apply to what you are doing, you MUST load and follow the skill.
In pi, use /skill:name to load a skill. Skill descriptions are in the system prompt — scan them before every task.
Load relevant skills BEFORE any response or action. Even a 1% chance means load the skill.
/skill:name, announce "Using [skill] to [purpose]"| Thought | Reality |
|---|---|
| "Just a simple question" | Questions are tasks. Check for skills. |
| "Need more context first" | Skill check comes BEFORE clarifying questions. |
| "Let me explore first" | Skills tell you HOW to explore. |
| "Too simple for a skill" | Simple things become complex. Use it. |
| "I remember this skill" | Skills evolve. Read current version. |
Use the subagent tool for delegated execution:
/implement <task> — scout → planner → worker chain/scout-and-plan <task> — scout → planner (no implementation)/implement-and-review <task> — worker → reviewer → worker/review <scope> — reviewer on recent changes/scout <query> — fast codebase recon