LLM & Agents

Synchronize MCP (Model Context Protocol) server settings across different Agent tools. Use when you want to copy, apply, or sync MCP settings between multiple Agent environments like Claude Code, VSCode, Claude Desktop, Codex, and Gemini CLI.

LangChain framework for LLM applications. Covers model wrappers (HuggingFace, Ollama), prompt templates, few-shot learning, output parsing, and chaining techniques for building sophisticated LLM workflows.

Design tools that agents can use effectively. Use when creating new tools for agents, debugging tool-related failures, or optimizing existing tool sets.

Homeostatic sensor validating test coverage percentage and detecting requirements without tests. Calculates coverage per requirement (REQ-*) and overall. Use as quality gate or continuous coverage monitoring.

Generates optimized prompts for AI logo generation tools (Midjourney, DALL-E, Stable Diffusion, ComfyUI). Transforms brand requirements into effective AI prompts with style parameters, negative prompts, and technical specifications.

Expert graphic, UI/UX designer for creating mockups, wireframes, and visual designs. Use when user needs help with designing interfaces, creating mockups, prototyping layouts, visual design feedback, design systems, brand identity, choosing colors/typography, or creating graphics. Triggers include "design," "mockup," "wireframe," "UI," "UX," "interface," "visual," "brand," or "prototype."

Use when user wants to set up comprehensive automation for their project. Generates custom subagents, skills, commands, and hooks tailored to project needs. Creates a multi-agent system with robust communication protocol.

Extract classes, functions, or code blocks from large modules into separate files. This skill should be used during refactoring to break up monolithic files while maintaining import compatibility and test coverage.

Use PROACTIVELY when creating specialized Claude Code sub-agents for task delegation. Automates agent creation following Anthropic's official patterns with proper frontmatter, tool configuration, and system prompts. Generates domain-specific agents, proactive auto-triggering agents, and security-sensitive agents with limited tools. Not for modifying existing agents or general prompt engineering.

Handles MMKV storage operations and data persistence patterns with encryption. Use when implementing data persistence, caching, or user preferences in Fitness Tracker App.

Create YouTube thumbnail/cover image prompts from video transcripts. Use when the user provides a transcript, script, or video content and wants thumbnail prompts for Nano Banana 3 or MidJourney. Generates click-optimized visuals and viral titles.

Generate a plan for how an agent should accomplish a complex coding task. Use when a user asks for a plan, and optionally when they want to save, find, read, update, or delete plan files in the repo .codex/plans.

Use when setting up project-specific agents via /cc:setup-project or when user requests custom agents for their codebase - analyzes project to create specialized, project-aware implementer agents that understand architecture, patterns, dependencies, and conventions

Master prompt injection attacks, jailbreak techniques, input manipulation, and payload crafting. Use when testing LLM prompt vulnerability or designing attack scenarios.

Build Chrome extensions with React, TypeScript, Tailwind CSS, and shadcn/ui. Use when users need to create browser extensions, content scripts, or popup UIs. Covers Manifest V3, messaging, storage, and permissions.

Design and implement memory architectures for agent systems. Use when building agents that need to persist state across sessions, maintain entity consistency, or reason over structured knowledge.

Expert at Playwright E2E tests, Vitest unit tests, Storybook interaction tests. Use when writing tests, debugging test failures, or improving test coverage.

LangChain high-level agent framework. Build agents with tools, memory, and streaming in under 10 lines of code.

Use when setting up Vercel projects, configuring blob storage, managing environment variables, or setting up custom domains with wildcard subdomains. Critical for avoiding env var loss and understanding wildcard routing requirements.

User interface animations and transitions. When Claude needs to create performant, tasteful animations.