ワンクリックで
mode-build
Runtime instructions for Build mode: implement changes, keep progress visible, test, and verify.
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
Runtime instructions for Build mode: implement changes, keep progress visible, test, and verify.
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
Runtime instructions for Ask mode: answer directly without code mutation.
Runs configurable multi-round research with compare, review, and explore modes, visible progress, source collection, and final sourced answers.
Creates and maintains project-board tasks with correct dedupe, umbrella/root linking, dependencies, tags, authors, assignments, and autopilot queue placement.
Runtime instructions for Plan mode: produce implementation or investigation plans without code mutation.
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".
| name | mode-build |
| description | Runtime instructions for Build mode: implement changes, keep progress visible, test, and verify. |
Implement the requested change by writing code, editing files, and running the smallest relevant verification.
#SLOPPY-12, or follows a Plan-mode task handoff, fetch the task details first with project.task_get or project.task_list.planning.progress_update with a compact checklist and a Definition of Done for each item.in_progress before working on it, mark it done only after concrete evidence or checks, mark it blocked with details when stuck, and use skipped when intentionally out of scope.agents.delegate_task only for independent, non-blocking side work.toolsets.ui-visual-verification skill as part of verification.