Browse skills

Expert guidance for Groq, the LLM inference platform that provides the fastest token generation speeds available, powered by custom LPU (Language Processing Unit) hardware. Helps developers integrate Groq's API for real-time AI applications where latency matters — chatbots, code completion, and streaming responses.

Design and execute growth experiments for rapid user acquisition and retention. Use when tasks involve viral loops, referral programs, conversion funnel optimization, A/B testing strategies, CAC/LTV analysis, product-led growth, activation rate improvement, cohort analysis, or scaling user growth through data-driven experimentation. Covers the full growth lifecycle from acquisition through retention.

Expert guidance for Grype, the open-source vulnerability scanner by Anchore that finds known vulnerabilities (CVEs) in container images, filesystems, and SBOMs. Helps developers integrate Grype into CI/CD pipelines, triage findings, and combine it with Syft for SBOM generation.

You are an expert in Guidance, Microsoft's library for controlling LLM output with constrained generation. You help developers write programs that interleave text generation with control flow (loops, conditionals, regex constraints, JSON schemas, function calls) — ensuring LLM output always matches the expected format by constraining the token generation process itself, not just prompting.

Run DOM tests without a browser using Happy DOM — fast JavaScript DOM implementation. Use when someone asks to "test without a browser", "Happy DOM", "fast DOM environment for tests", "replace jsdom", "Vitest DOM environment", or "server-side DOM". Covers test environment setup, Vitest/Jest integration, SSR testing, and performance comparison with jsdom.

Assists with managing secrets, encryption keys, and dynamic credentials using HashiCorp Vault. Use when configuring secret engines, setting up dynamic database credentials, implementing access policies, integrating with Kubernetes, or using Transit for encryption as a service. Trigger words: vault, secrets management, dynamic secrets, transit engine, pki, approle, vault agent.

Orchestrate background jobs and workflows with Hatchet — open-source distributed task queue with DAG workflows. Use when someone asks to "run background jobs", "Hatchet", "workflow orchestration", "distributed task queue", "durable execution", "replace Celery/Bull", or "DAG workflow engine". Covers workflow definition, step functions, retries, concurrency control, and event-driven triggers.

LLM observability proxy that sits between your app and LLM providers. Logs every request, enables caching, rate limiting, and provides cost analytics. Works with OpenAI, Anthropic, and other providers with a one-line integration change.

Manage Hetzner Cloud infrastructure from the terminal. Use when a user asks to create a Hetzner server, manage VPS instances, set up firewalls, configure networks, manage volumes, create snapshots, handle SSH keys, or provision infrastructure on Hetzner. Covers the hcloud CLI for all resource types. For deploying applications on top of Hetzner servers, see coolify.

Implement HIPAA compliance for healthcare applications — PHI handling, encryption, audit logging, access controls, and Business Associate Agreements. Use when building healthtech apps, handling patient data, preparing for a HIPAA audit, or signing hospital/payer contracts.

Expert in Zod — TypeScript-first schema validation. Covers parsing, custom errors, refinements, type inference, and integration with React Hook Form, Next.js, and tRPC.

You are an expert in Hotwire (HTML Over The Wire), the collection of techniques for building modern web applications by sending HTML instead of JSON. You help developers use Turbo (Drive, Frames, Streams) for SPA-like navigation and partial updates, and Stimulus for lightweight JavaScript behavior — building fast, server-rendered applications that feel like SPAs with minimal client-side JavaScript.

htmx gives you access to AJAX, CSS transitions, WebSockets, and Server-Sent Events directly in HTML using attributes. It enables dynamic web UIs without writing JavaScript by letting any element issue HTTP requests and swap content into the DOM.

Work with Hugging Face's ecosystem for machine learning — transformers library, model hub, tokenizers, inference pipelines, and fine-tuning. Covers downloading pre-trained models, running inference, training custom models, and publishing to the Hub.

Run scripts on Git hooks with Husky. Use when a user asks to run linters before commit, validate commit messages, run tests before push, or set up Git hooks for a team.

Reference for US International Building Code (IBC) occupancy types, construction types, height/area limits, egress requirements, and R-2 residential standards. Use when: generating US-compliant building models, validating building designs against IBC, understanding occupancy classifications and construction types.

Access macOS native data (Messages, Contacts, Reminders, Calendar) via MCP server with iMCP. Use when: building AI assistants that access Apple data, querying contacts/reminders from Claude Desktop, integrating macOS apps with AI agents.

Set up and manage InfluxDB for time-series data storage, querying, and analysis. Use when a user needs to configure InfluxDB buckets, write Flux queries, set up retention policies, create tasks for data downsampling, or build dashboards for time-series metrics.

Create, optimize, and automate Instagram content including Reels, Stories, carousels, and feed posts. Use when someone asks to "grow on Instagram", "create Instagram Reels", "Instagram content strategy", "Instagram API integration", "automate Instagram posting", "Instagram analytics", or "Instagram marketing". Covers Reels-first strategy, visual content guidelines, Instagram Graph API, and growth tactics.

Add customer messaging and support with Intercom. Use when a user asks to add in-app messaging, build a help center, set up customer onboarding flows, create product tours, or implement a support ticketing system.