grill-me
Use when the user wants a plan or design stress-tested before building it. Triggers on "grill me", "grill this plan", "stress-test this plan", "poke holes in this design", "interrogate this design/plan", AND on pre-implementation moments where a concrete plan or design is on the table and about to be implemented. Dispatches a separate read-only grill agent that interrogates the plan one question at a time; the main agent answers each question from the codebase and escalates only costly-to-reverse decisions (schema/contract, public API, data migration, security boundary) to the human.
What this skill does
# grill-me
Stress-test a plan or design before implementation by running a relentless, one-question-at-a-time interrogation through a separate **grill agent**. The grill agent asks; YOU (the main agent) answer — usually from the codebase; the human is pulled in only for costly-to-reverse decisions.
## When this applies
- **Explicit:** "grill me", "grill this plan", "stress-test this plan", "poke holes in this design", "interrogate this design".
- **Implicit:** a concrete plan or design is on the table and about to be built, and it should be vetted first.
If there is no concrete plan/design yet, say so and offer to brainstorm one first — there is nothing to grill.
## The relay loop
This loop is the whole skill. Run it.
1. **Spawn the grill agent once.** Use the Agent tool with:
- `subagent_type: "grill"`
- prompt: the plan/design under review verbatim, plus pointers to the relevant files/dirs, ending with "Ask your first question."
- Leave permission mode at default — the grill agent is read-only and needs no edit permissions.
- **Capture the `agentId` from the spawn result** (format `a…`; the result reads "use SendMessage with to: '<agentId>'"). You address the grill agent by this `agentId` for every subsequent round — a fresh `Agent` call would start over with no memory.
2. **Receive one question** — a `QUESTION / WHY / RECOMMENDED` block — or `DONE`. **If the response is `DONE`, skip straight to "On completion" below** — do not log it as a round and do not send another answer.
3. **Answer it yourself.** Derive the answer from the codebase, project conventions, git history, or sound engineering defaults; Read/grep as needed.
- **Escalate to the human ONLY** when the question concerns a costly-to-reverse decision: a schema/contract change, a public API shape, a data migration, or a security boundary. For those, use `AskUserQuestion`.
- For everything else, answer autonomously — regardless of your confidence.
4. **Log the round.** Append one row to a running decision log you keep visible in the conversation:
| # | Question | Answer | Source |
|---|----------|--------|--------|
`Source` is one of: `codebase`, `convention`, `git`, `default`, `HUMAN`.
5. **Continue the grill agent** with the answer, addressing it by the captured `agentId`:
- `SendMessage({ to: "<agentId>", message: "ANSWER <n>: <your answer>" })`
- `<n>` is the round number from the `QUESTION <n>` you just received.
- It returns the next question, or `DONE`.
6. **Stop when** the grill agent emits `DONE` (handled at step 2). Independently, if you reach the safety cap of **15 rounds** (use a different number only if the user named one), pause and tell the user the cap was reached. If they say "keep going", raise the cap by another 15 and continue the loop — the grill agent is still active and needs no re-spawn. Otherwise, send `SendMessage({ to: "<agentId>", message: "CAP REACHED — summarize unresolved threads and emit DONE" })` and use its summary.
## On completion
1. Print the final decision log (the fully populated table above).
2. **Revise the plan/design on the table** to fold in every resolved decision, so the user leaves with an updated plan — not just a transcript. Show the revised plan or a diff of what changed.
3. List any open items explicitly: anything the human deferred, or threads the cap cut short.
## Guardrails
- The grill agent is **read-only** — it cannot change files. All edits happen in this main session.
- One question per round. Do not let the loop batch questions or skip the answer step.
- Keep answers grounded: name the file or convention you used in the `Source` column.
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.