Loading...
Loading...
Found 60 Skills
Specification Generator — Generate project requirements, design documents, and task lists. A skill for generating structured project specifications through interactive dialogue or quick generation. English triggers: - "Create requirements", "Generate requirements doc", "Summarize as requirements" - "Create design document", "Design the architecture", "Generate technical spec" - "Create task list", "Break down into tasks", "Generate tasks.md" - "Create full spec", "Generate all specs", "Create the complete specification" - After discussion: "Turn this into requirements", "Document this as spec" 日本語トリガー: - 「要件定義を作って」「要件をまとめて」「仕様書を作成して」 - 「設計書を作って」「技術設計をして」「アーキテクチャを設計して」 - 「タスクリストを作って」「実装タスクに分解して」「tasks.mdを生成して」 - 「仕様を全部まとめて」「フル仕様を作成」「3点セットを作って」 - 会話で仕様が固まった後に「これを要件定義書にして」
Structured TODO commit workflow using JJ (Jujutsu). Use to plan tasks as empty commits with [task:*] flags, track progress through status transitions, manage parallel task DAGs with dependency checking. Enforces completion discipline. Enables to divide work between Planners and Workers. **Requires the working-with-jj skill**
Add todos to the current plan to prevent mistakes.
Use when converting architecture or design documents into structured, dependency-ordered implementation task lists for autonomous agent execution via dark-factory
[Docs] ⚡⚡⚡⚡ Analyze the codebase and create initial documentation
Analyze multi-step tasks to identify which steps can run in parallel, build dependency graphs, detect conflicts (write-write, read-write, resource contention), and produce optimized execution plans. Triggers on 3+ independent steps, "speed up", "run simultaneously", "parallelize", "optimize" or any task where sequential execution wastes time.
Make today's plan from my Plane tasks.
Decompose technical design into agent-sized implementation issues → numbered markdown files. Triggers: 'plan this,' 'break into issues,' 'create tasks,' 'ready to implement,' post-architect. Not for: designs without file paths/phases (run architect first).
Internal conversation-entry router for Claude Code. Performs a lightweight intake pass at the start of substantive work: decide whether a more specific skill should be invoked first, whether specialized agent delegation is warranted, how much context is actually needed, and whether the task needs planning or can proceed directly. Not a user-facing slash command.
Plans development tasks in a structured way for teams. Use this skill when the user triggers /plan-task with a description, or when asking to plan, detail, or break down a development task into steps. Always use when the message starts with /plan-task.
Превращает идею в конкретную задачу, которую агент может выполнить. Проводит через 5 шагов: идея → результат → реальность → задача → план.
Breaks work into bite-sized tasks before coding. Activates when a multi-step task needs planning — creates tasks small enough for a junior developer to follow (2-5 minutes each), with exact file paths, complete implementation details, and verification steps. References Linear issue context and project-specific test commands from CLAUDE.md.