plan
Plan Mode workflow. Use this skill when the user says "/plan", "help me plan", "break this into tasks", "design the approach", or otherwise wants to plan work, and also whenever you are in Plan Mode. Drives all planning activity — research, task decomposition, and creating kanban tasks as the plan artifact.
What this skill does
# Plan
Use whenever you enter Plan Mode or the user asks you to plan work.
$ARGUMENTS
{% include "_partials/delegate-to-subagent" %}
## Interpreting the arguments
The arguments above may be a free-form description of the work, a path to a file that is the basis for the plan, or both.
- **If the arguments name or reference a file** (e.g. a path like `docs/spec.md`, an `@`-mention, or "plan from <file>"), read that file first with `Read` and treat its contents as the authoritative basis for the plan. Follow any further file references inside it that are relevant.
- **If the arguments are a description**, plan from the description directly.
- **If both are given**, the description refines or scopes what's in the file.
Either way, still do the `code_context` research below before creating tasks — the basis file tells you *what* to build; research tells you *what's affected*.
## Goals
1. **Understand the work** — research deeply enough to know what changes and what's affected.
2. **Produce a kanban board** — the artifact is kanban tasks with subtasks. Not markdown. Not TodoWrite/TaskCreate/TaskUpdate.
3. **Right-size tasks** — each is one focused unit, independently implementable and verifiable.
4. **Collaborate** — present, discuss, iterate until the user is satisfied.
5. **Hand off cleanly** — when done, remind the user: `/finish` (autonomous) or `/implement` (one at a time).
## Example
**Feature request → decomposed board:** User says "I want to add authentication to the app".
1. Research with `code_context`: `search symbol "user"`, `search symbol "session"`, `get blastradius src/server.rs max_hops 3` to find boundaries and callers.
2. Ensure a board exists: `kanban` `{"op": "init board", "name": "<repo name>"}` (`add task` auto-creates one, but name it).
3. As design crystallizes in conversation, create tasks one at a time with the `kanban` tool — not as an end-of-discussion batch. Each `description` follows the Task Standards template (What / Acceptance Criteria / Tests):
- `{"op": "add task", "title": "Design auth architecture", "description": "## What\n…\n## Acceptance Criteria\n- [ ] …\n## Tests\n- [ ] …"}`
- `{"op": "add task", "title": "Add User model and migration", "description": "## What\n…"}`
- `{"op": "add task", "title": "Implement POST /api/login", "description": "…", "depends_on": ["<user-model-task-id>"]}`
4. Encode ordering with `depends_on` so foundational tasks precede integration.
5. Verify with `{"op": "list tasks"}`, present the board, iterate.
6. User approves → remind: `/finish` (autonomous) or `/implement` (one at a time). Do NOT call `ExitPlanMode`, do NOT start implementing.
The board IS the plan. **Never write a markdown plan file** (`PLAN.md`, `DRAFT_PLAN.md`, scratch files) — `/finish` and `/implement` read kanban, not prose. If the `kanban` tool is unavailable or its calls fail, STOP and tell the user; do not substitute markdown and do not claim tasks exist without a `list tasks` read-back.
## Constraints
{% include "_partials/architecture-awareness" %}
### Plans are kanban tasks — created as you go
Every planned item becomes a kanban task. The board IS the plan; no markdown files. **Create tasks as they crystallize during discussion, not at the end.** If a work item is defined enough to describe in conversation, it's defined enough to be a task. Don't wait to be asked.
### Research before tasks
`code_context` is primary. Always run `get blastradius` on files you expect to change — that's how you find downstream work you'd otherwise miss. Use symbol search, callgraphs, and text search (Glob/Grep/Read) to fill in the picture.
{% include "_partials/task-standards" %}
### Board naming
Name the board for the workspace/repository, not the feature being planned.
### User controls plan-mode exit
Do NOT call `ExitPlanMode`. The user decides when the plan is ready.
### No auto-implementation on exit
When the user exits plan mode or approves, do NOT begin implementing. Remind:
- `/finish` — drives tasks to `done` (implement → test → review) autonomously
- `/implement` — one task at a time
### Ordering
Foundational changes (data models, types, config) → core logic → integration → tests → cleanup. Use `depends_on` for ordering constraints.
## Autonomous Agent Mode
No Plan Mode UI or TUI (e.g. headless `-p`)? The procedure above is unchanged: research, then create kanban tasks one at a time with the `kanban` tool, and verify with `list tasks`. Do not wait for a UI and do not write a markdown plan file. (`references/PLANNING_GUIDE.md` has the long-form version when bundled, but everything required is in this file.)
## Updating an Existing Plan
Update kanban directly — add tasks, `update task` to edit, `delete task` to remove, reorder dependencies. The board is a living document.
Related in Design
contribute
IncludedLocal-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers at ~/.contribute-system/research/, append-only event log at ~/.contribute-system/log.jsonl. No database, no cloud calls. Use when the user asks about their PRs / issues / contributions, wants to find new work to take on, claim an issue, build/refresh a repo's dossier, or draft a Design Issue or PR. Trigger with "/contribute", "what's my PR status", "find a contribution", "claim issue X", "draft a Design Issue for Y", "refresh dossier for Z".
architectural-analysis
IncludedUser-triggered deep architectural analysis of a codebase or scoped subtree across eight modes — information architecture, data flow, integration points, UI surfaces, interaction patterns, data model, control flow, and failure modes. This skill should be used when the user asks to "diagram this codebase," "map the architecture," "show the data flow," "give me an ERD," "trace control flow," "find the integration points," "verify the layout pattern," "audit the UX architecture," or any similar request whose primary deliverable is mermaid diagrams plus cited reports under docs/architecture/. Dispatches haiku/sonnet sub-agents in parallel for per-mode exploration, then verifies every citation mechanically before any node lands in a diagram. Not for one-off prose explanations of code (use code-explanation) or for high-level system design from scratch (use system-design).
mcp
IncludedModel Context Protocol (MCP) server development and tool management. Languages: Python, TypeScript. Capabilities: build MCP servers, integrate external APIs, discover/execute MCP tools, manage multi-server configs, design agent-centric tools. Actions: create, build, integrate, discover, execute, configure MCP servers/tools. Keywords: MCP, Model Context Protocol, MCP server, MCP tool, stdio transport, SSE transport, tool discovery, resource provider, prompt template, external API integration, Gemini CLI MCP, Claude MCP, agent tools, tool execution, server config. Use when: building MCP servers, integrating external APIs as MCP tools, discovering available MCP tools, executing MCP capabilities, configuring multi-server setups, designing tools for AI agents.
react-native-skia
IncludedDesign, build, debug, and optimise high-polish animated graphics in React Native or Expo using @shopify/react-native-skia, Reanimated, and Gesture Handler. Use when the user wants canvas-driven UI, shaders, paths, rich text, image filters, sprite fields, Skottie, video frames, snapshots, web CanvasKit setup, or performance tuning for custom motion-heavy elements such as loaders, hero art, cards, charts, progress indicators, particle systems, or gesture-driven surfaces. Also use when the user asks for fluid, glow, glass, blob, parallax, 60fps/120fps, or GPU-friendly animated effects in React Native, even if they do not explicitly say "Skia". Do not use for ordinary form/layout work with standard views.
plaid
IncludedProduct Led AI Development — guides founders from idea to launched product. Six capabilities: Idea (discover a product idea), Validate (pressure-test the idea against fatal flaws, problem reality, competition, and 2-week MVP feasibility), Plan (vision intake + document generation), Design (translate image references into a design.md spec), Launch (go-to-market strategy), and Build (roadmap execution). Use when someone says "PLAID", "plaid idea", "help me find an idea", "product idea", "idea from my business", "idea from my expertise", "plaid validate", "validate my idea", "pressure-test", "is this idea good", "find fatal flaws", "validate the problem", "plan a product", "define my vision", "generate a PRD", "product strategy", "plaid design", "design from image", "translate image to design", "create design.md", "extract design tokens", "plaid launch", "go-to-market", "launch plan", "GTM strategy", "launch playbook", "plaid build", "build the app", "start building", or "execute the roadmap".
nextjs-framer-motion-animations
IncludedAdds production-safe Motion for React or Framer Motion animations to Next.js apps, including reveal, hover and tap micro-interactions, whileInView, stagger, AnimatePresence, layout and layoutId transitions, reorder, scroll-linked UI, and lightweight route-content transitions. Use when the user asks to add, refactor, or debug Motion or Framer Motion in App Router or Pages Router codebases, especially around server/client boundaries, reduced motion, LazyMotion, bundle size, hydration, or route transitions. Avoid for GSAP-style timelines, WebGL or 3D scenes, heavy scroll storytelling, or CSS-only effects unless Motion is explicitly requested.