ワンクリックで
executing-plans
Use when you have a written implementation plan to execute in a separate session with review checkpoints
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 you have a written implementation plan to execute in a separate session with review checkpoints
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。
Scan design/ for backlog/todo artifacts, check if work has been merged into the codebase, and auto-rename completed ones to done-. Use for periodic hygiene, after merging branches, or to audit design artifact status.
Use when writing Dagger pipelines in Python, building containerized CI/CD with the Dagger Python SDK, creating Dagger modules/functions/objects, integrating services (PostgreSQL, Redis) in Dagger, using Dagger LLM/agent features, or running Dagger in GitHub Actions.
Use when creating specifications, implementation plans, or tools for a feature - enforces consistent naming across related artifacts
Use this skill when the user wants to run hotdata CLI commands, query the Hotdata API, list workspaces, list connections, create connections, list tables, manage datasets, execute SQL queries, inspect query run history, search tables, manage indexes, manage sandboxes, manage workspace context and the data model via the context API (`hotdata context`), or interact with the hotdata service. Activate when the user says "run hotdata", "query hotdata", "list workspaces", "list connections", "create a connection", "list tables", "list datasets", "create a dataset", "upload a dataset", "execute a query", "search a table", "list indexes", "create an index", "list query runs", "list past queries", "query history", "list sandboxes", "create a sandbox", "run a sandbox", "workspace context", "pull context", "push context", "data model", or asks you to use the hotdata CLI.
Use when building, deploying, or debugging applications on Modal (modal.com) — serverless cloud compute for Python. Covers container images, GPU functions, web endpoints, volumes, secrets, scheduling, distributed training, calling deployed functions, and scaling patterns.
Use when calling deployed Modal functions from a CLI or external service without needing the Modal app running locally
SOC 職業分類に基づく
| name | executing-plans |
| description | Use when you have a written implementation plan to execute in a separate session with review checkpoints |
| hooks | {"PreToolUse":[{"matcher":"Write|Edit|Bash","hooks":[{"type":"command","command":"R=$(dirname \"$(git rev-parse --git-common-dir 2>/dev/null)\" 2>/dev/null); f=$(ls \"$R\"/design/{todo,backlog}-*-plan-*.md 2>/dev/null | head -1); [ -f \"$f\" ] && head -30 \"$f\"; true"}]}],"Stop":[{"hooks":[{"type":"command","command":"R=$(dirname \"$(git rev-parse --git-common-dir 2>/dev/null)\" 2>/dev/null); f=$(ls \"$R\"/design/{todo,backlog}-*-plan-*.md 2>/dev/null | head -1); [ -f \"$f\" ] && { DONE=$(grep -c \"^- \\[x\\]\" \"$f\" || echo 0); TODO=$(grep -c \"^- \\[ \\]\" \"$f\" || echo 0); echo \"[executing-plans] $DONE steps done, $TODO remaining.\"; [ \"$TODO\" -gt 0 ] && echo \"[executing-plans] Open tasks remain — verify completion before stopping.\"; }; true"}]}]} |
Load plan, review critically, execute all tasks, report when complete.
Announce at start: "I'm using the executing-plans skill to implement this plan."
Note: If subagents are available, prefer subagent-driven-development for faster iteration. Use this skill when you explicitly want inline or batched execution in one session.
systematic-debugging before proposing fixes.superpowers: prefixes).This skill defines hooks that mechanically prevent goal drift during long execution sessions:
These hooks find plan files in the primary checkout's design/ directory automatically, including from worktrees.
If resuming plan execution after /clear, a new session, or context compression, answer these five questions before continuing:
| Question | Source |
|---|---|
| Where am I? | Current task checkbox in plan file |
| Where am I going? | Remaining unchecked tasks |
| What's the goal? | Plan header **Goal:** line |
| What have I learned? | git log + any findings from prior work |
| What have I done? | git diff main...HEAD --stat |
If you can answer all five, proceed from the current task. If not, re-read the plan file and git history until you can.
For each task:
After all tasks complete and verified:
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to Review (Step 1) when:
Don't force through blockers - stop and ask.
Required workflow skills: