runiondd

Design the user interface and user experience for a product based on its PRD and architecture. Use this skill when someone says "design the UI", "what should this look like", "wireframes", "layout", "user experience", "UX", "screen design", "component design", "design system", or anything about turning requirements into visual interface specifications. Trigger after the PRD is complete and ideally after architecture, so you know what data and actions are available.

Translate a PRD into technical architecture decisions and an ordered task breakdown for autonomous building. Use this skill after a PRD is complete, when someone says "let's figure out the tech stack", "architect this", "technical design", "break this into tasks", "system design", "how should we build this", or anything about turning product requirements into technical decisions and implementation plans. Also trigger when reviewing or refining an existing architecture document.

Create deployment runbooks, infrastructure plans, and CI/CD configurations. Use this skill when someone wants to deploy a project, set up hosting, create a deployment pipeline, configure CI/CD, write a runbook, plan infrastructure, or says things like "how do we deploy this", "get this to production", "set up hosting", "deployment checklist", "infrastructure plan", or "DevOps setup". Also trigger when reviewing or improving an existing deployment process.

Create go-to-market strategy and launch assets including positioning, messaging, landing page copy, email campaigns, and launch plans. Use this skill when someone mentions GTM, go-to-market, marketing plan, launch strategy, product launch, landing page copy, email campaign, messaging framework, value proposition, competitive positioning, or says things like "how do we launch this", "marketing materials", "tell the world about this", "get users", "growth strategy", or "brand messaging". Trigger even for casual mentions like "we need to market this thing" or "how do we get customers".

Define analytics tracking plans, monitoring strategies, KPIs, dashboards, and alerting configurations. Use this skill when someone wants to set up analytics, monitoring, observability, tracking, dashboards, KPIs, metrics, alerts, or says things like "how do we measure this", "what should we track", "set up monitoring", "analytics plan", "instrument the app", "key metrics", or "how do we know if this is working". Also trigger when reviewing an existing instrumentation setup or when someone wants to improve their observability.

Generate user documentation, internal SOPs, knowledge base articles, troubleshooting guides, and onboarding materials. Use this skill when someone wants documentation, help articles, SOPs, runbooks, knowledge base content, user guides, onboarding docs, FAQs, troubleshooting guides, or says things like "write the docs", "document this", "user guide", "help center content", "how-to articles", "support documentation", "internal procedures", or "operations manual". Trigger even for casual mentions like "we need docs for this" or "customers will need help with this".

Generate a comprehensive test plan from a PRD and architecture document. Use this skill when someone asks for a test plan, test strategy, QA plan, wants to define how to test a product, or says things like "how should we test this", "write tests for", "test coverage", "QA strategy", or "testing approach". Also trigger when reviewing an existing test plan or when someone wants to ensure their acceptance criteria are thorough enough for autonomous testing.

Kick off any new product or project by producing a complete, build-ready PRD with machine-readable acceptance criteria. Use this skill whenever starting a new project, kicking off a build, beginning product planning, or when someone says "let's start a new project", "I have an idea for...", "new product", "new feature", "kick this off", "get this started", or anything suggesting they want to go from idea to actionable spec. This is the entry point for the Product Launch Framework. Even if the user already has a PRD, use this skill to validate it has acceptance criteria tight enough for autonomous building.

Use when the user types /interactive-learning or says "make it interactive" or "visualize this lesson" — to turn a learn/<slug>.md lesson into a single self-contained HTML page with a control (slider/toggle/input) that makes the lesson's mechanism obvious by playing with it. Reads from the lesson file, so it works in any chat or repo. Pairs with the learn skill.

Use when the user types /learn or says "extract the lesson", "capture the lesson", or "what did we learn" after finishing a study, bug fix, or feature — to distill a piece of work into one reusable lesson file at learn/<slug>.md in the current repo. The work can come from an explicit topic, a file or diff pointed at, recent git activity, or the conversation. Output is a single concept future-you can reread cold.

Use when deciding whether to delegate a task to a cheaper-model subagent for cost efficiency — code/symbol location, log scans, mechanical renames, boilerplate, writing tests, focused diff review, or single-file edits. Provides the break-even gate, the haiku/sonnet/opus routing taxonomy, and hard delegation guardrails. Always-on cost discipline.