schema-author
Design the Zod output schema of a Smithers <Task> as the contract between steps. Use when a step's output feeds a later step (or a branch/loop condition) and must be reliable — design the schema first, keep it minimal, and prefer typed fields over prose so downstream rendering can depend on it.
What this skill does
# Schema Author
This skill is about **one thing: the output schema** — the Zod shape a `<Task>`
produces and the next step consumes. In Smithers, that schema *is* the contract.
The runtime injects a JSON-schema description of it into the prompt, parses the
agent's response, validates against Zod, retries on mismatch, and persists the
row. Everything downstream — a `ctx.outputMaybe(...)` conditional, a `<Branch>`,
a `<Loop until={...}>` — reads that row. A loose or prose-heavy schema makes every
later step unreliable; a tight, typed one makes the graph deterministic.
This is the BAML insight: **the prompt is a schema.** You don't beg the model for
JSON in prose, you declare the type and let the runtime enforce it. Design the
contract before you write the prompt or the workflow.
## When to reach for it
- A step's output is read by a *later* step, a branch condition, or a loop's
`until`, and a wrong shape would silently break the run.
- An agent keeps returning the right *idea* in the wrong *shape* (free text where
you need an enum, missing a field downstream code indexes into).
- You're about to add a reviewer/retry to compensate for output you could just
*type* instead.
Skip it when the output is terminal (nothing downstream reads it) — a `summary`
string is fine. Schema rigor is for fields other steps depend on.
## Design the contract first, keep it minimal
Author the schema in `createSmithers({...})` *before* the prompt or the graph.
Include only what downstream actually reads — a one-line `summary` plus the few
fields the next step indexes into. Every extra field is another thing the agent
can get wrong and another retry.
```tsx
const { Workflow, smithers, outputs } = createSmithers({
triage: z.object({
summary: z.string(), // human-readable, terminal
severity: z.enum(["low", "medium", "high"]), // a <Branch> reads this
category: z.enum(["bug", "feature", "question"]), // routes to a specialist
needsHuman: z.boolean(), // gates an <Approval>
}),
});
```
- **Prefer enums and typed fields over prose.** `z.enum([...])`, `z.boolean()`,
`z.number()` give the next step something it can switch on. A free-string status
is a bug waiting for a typo.
- **Make required things required.** Optional fields the downstream step assumes
exist are the classic silent failure. If `fix` always reads `analysis.issues`,
don't make `issues` optional.
- **Constrain values, not just types.** `z.number().min(0).max(100)`,
`z.array(...).min(1)` — a validation failure feeds the error back and the agent
self-corrects on retry, so tighter bounds are free reliability. Annotate
non-obvious fields with `.describe("...")`; that text rides into the injected
JSON-schema block and steers the agent.
## Wire it: every `<Task>` gets `output={outputs.x}`
The schema is referenced by the typed `outputs.x` handle, which gives compile-time
checks (a typo in the key is a type error):
```tsx
<Task id="triage" output={outputs.triage} agent={analyst}>
{`Triage: ${ctx.input.report}`}
</Task>
{/* downstream reads typed fields — no string parsing, no guessing */}
<Branch
if={ctx.outputMaybe(outputs.triage, { nodeId: "triage" })?.severity === "high"}
then={<Task id="escalate" .../>}
else={<Task id="queue" .../>}
/>
```
The prompt body stays clean: end it with the task, let the runtime append the
schema. Don't hand-write a "return JSON like {…}" block — it fights the injected
one (see `skills/prompt-author/SKILL.md`).
## Rich or extensible outputs: `z.looseObject`
When you can't enumerate every field up front (a typed-extraction step, a payload
that carries pass-through metadata, an evolving spec), use `z.looseObject({...})`:
name and type the fields downstream *depends on*, and let the agent attach extra
keys without tripping validation. You keep a reliable contract on the load-bearing
fields and an open door for the rest.
```tsx
extract: z.looseObject({
title: z.string(),
amount: z.number(), // downstream math reads this
// agent may also return vendor, date, lineItems… — preserved, not rejected
}),
```
Use a strict `z.object` when the shape is a true contract a branch/loop keys off;
use `z.looseObject` when richness and forward-compatibility matter more than
locking the shape.
## Verify the contract holds
Attach a `schemaAdherence` scorer to confirm the shape holds run to run, and read
the persisted row directly:
```bash
bunx smithers-orchestrator scores <run-id> # did adherence hold?
bunx smithers-orchestrator output <run-id> triage # see the persisted row
```
See `skills/smithers/SKILL.md` for the runtime/CLI surface and `docs/llms-core.txt`
("The runtime injects a JSON-schema description … validates against Zod") for the
exact validate-and-retry mechanics.
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.