Use when the user supplies a natural-language UI component request that needs to be turned into a precise component specification. Fills the spec by inference, asking the director a question only for genuine UI branches that pass Principle 2's gate (in practice the interaction-model variant). Writes a Markdown spec to docs/gyre/specs/components/<kebab-name>.md. Component-only in this iteration; classifier recognizes page and design-guide requests but the skill terminates without writing a file in those cases.
Use when the user wants to generate a working custom UI component file plus a Storybook story file from an existing component request specification produced by clarify-component. Reads the spec at docs/gyre/specs/components/<kebab-name>.md, discovers host context (existing catalog, component output alias, primitive location, dependencies, Storybook setup), decides Reuse/Extend/Compose/New from that discovery (Principle 4), presents a Plan for approval, then writes the two files, audits them against the loaded stack skills' rules (self-review), and verifies them via typecheck, scoped lint, and Storybook build. Generates for New and Compose; for Extend and Reuse it reports the decision and terminates without generating (those paths are deferred). Invoked as /generate-component <kebab-name>.
Use when the user wants to prepare a host project for the AI Harness's component generation by installing the external stack skills it depends on. Reads the stack-skills manifest, detects which skills are already installed in the host project, installs the missing ones via the skills CLI, and reports the result. Invoked as /setup with no arguments.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
AI driven React and TypeScript code generation from natural language.
Frontend development typically routes through three roles — product manager, designer, and developer — and the back-and-forth between them is slow and prone to conflict driven by differing perspectives. When that conflict is healthy it lifts quality; when it isn't, it just burns energy.
Gyre replaces that structure with a single director ↔ AI (acting as PM, designer, and developer). The quality lift that used to come from human-to-human dialogue now has to be delivered by human-to-AI dialogue.
Given a natural-language request for a UI component or page, Gyre refines the under-specified intent through dialogue and produces one of:
DESIGN.md).Stable execution of this behavior marks completion. Beyond that point, the goal is to systematically manage and extend the component library while the AI keeps learning and evolving.
/setup from the host project./clarify-component "<natural-language request>" — a short natural-language request from the director.docs/gyre/specs/components/<kebab-name>.md (spec) — it captures the goal and behavior, not the implementation. The implementation is decided in step 2 (Generate)./generate-component <kebab-name> — the spec from step 1.Gyre is a Component-Driven Development (CDD) harness that realizes each Rung of the CDD altitude ladder from the bottom up. Current support by Rung:
| Rung | Type | Status |
|---|---|---|
| Primitive | — | 🟢 Supported |
| Composed | molecule — Compose from primitives | 🟢 Supported |
| organism — composed from higher-level components | 🔴 Not yet | |
| Pages | — | 🔴 Not yet |
npx claudepluginhub calelix/gyre --plugin gyreUltra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Frontend design skill for UI/UX implementation
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Memory compression system for Claude Code - persist context across sessions
Marketing skills for AI agents — conversion optimization, copywriting, SEO, paid ads, ad creative, and growth
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.