ワンクリックで
starting-point
Use when an outcome is stated, inferred, or hidden.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Use when an outcome is stated, inferred, or hidden.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | starting-point |
| description | Use when an outcome is stated, inferred, or hidden. |
| license | MIT |
| metadata | {"author":"Kiren Srinivasan","version":"0.1.0"} |
Start from the user's observable end state. Treat a prescribed method as a candidate path unless that method is the result or a fixed constraint. Never replace the user's priorities with your own.
Before acting, identify five things privately:
For complex work, define smaller outcomes whose evidence jointly proves the parent outcome. A component pass is not proof that the whole result is complete. Infer outcomes only from conversation evidence. Do not promote a weak inference, preference, or activity metric.
Keep requested work when it still contributes. Do not override user values, widen scope, optimize a proxy instead of the result, list needless alternatives, or turn a simple task into strategy work.
Use the proof checklist before claiming a result whose subtasks, approvals, or external effects can be confused with whole-outcome completion.
Once the route is selected, use the task-specific workflow for design, planning, debugging, testing, research, or execution. This skill chooses the starting boundary. It does not replace the method that performs the work.