メニュー
My planning skill that has additional instructions. Always use when entering plan mode.
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
Lint prose against Simplified Technical English (ASD-STE100) rules. Use when the user asks to check, lint, or simplify technical writing (specs, manuals, procedures, docs) for STE compliance, or mentions ASD-STE100 or Simplified Technical English.
Any time committing, pushing, or creating new revisions is mentioned, use this commit flow.
Step-by-step procedure for building a Claude Code channel integration. Use when creating a new channel plugin that connects Claude Code to external systems via MCP, named pipes, webhooks, or chat platforms.
This skill should be used when the user asks to "set up a dev environment", "create a workspace", "open a browser pane", "split terminal", "spin up frontend and backend", "use cmux", "manage cmux panes", "run commands in cmux", or mentions cmux workspaces, panes, surfaces, or browser automation. Provides workflow guidance for orchestrating multi-pane development environments with terminal and browser surfaces.
Use when merging a stack of stacked PRs with jj. Triggered by phrases like "merge the PR stack", "merge the stacked PRs", "land these PRs", or "merge the stack into main".
Use when addressing PR review comments on stacked jj branches. Triggered by phrases like "address review comments", "fix PR feedback", "respond to review on stacked branch", or "update the PR with reviewer suggestions".
| name | Plan |
| description | My planning skill that has additional instructions. Always use when entering plan mode. |
Prefer the simplest solution first. Do not over-engineer with new abstractions, polymorphic tables, or extra layers unless the user explicitly requests them. When in doubt, ask before adding complexity.
Know the limits of your skill and knowledge. If the acceptance criteria or error seems unclear, suggest an exploration and/or reproduction style session.
A dedicated subagent can and should be spun up to write a script to validate when appropriate. This should be done in parallel to the implementation.
