LLM & Agents

LLM function calling and tool use patterns. Use when enabling LLMs to call external tools, defining tool schemas, implementing tool execution loops, or getting structured output from LLMs.

Reviews Deep Agents code for bugs, anti-patterns, and improvements. Use when reviewing code that uses create_deep_agent, backends, subagents, middleware, or human-in-the-loop patterns. Catches common configuration and usage mistakes.

Execute feature plans from ~/.ai/plans using progressive disclosure and state tracking. This skill reads existing feature plans, tracks implementation progress in implementation_state.md, uses the Plan agent to explore plans and codebases, and systematically implements features phase by phase. Pairs with the feature-planning skill.

Text embeddings for semantic search and similarity. Use when converting text to vectors, choosing embedding models, implementing chunking strategies, or building document similarity features.

World-class ML engineering skill for productionizing ML models, MLOps, and building scalable ML systems. Expertise in PyTorch, TensorFlow, model deployment, feature stores, model monitoring, and ML infrastructure. Includes LLM integration, fine-tuning, RAG systems, and agentic AI. Use when deploying ML models, building ML platforms, implementing MLOps, or integrating LLMs into production systems.

Local LLM inference with Ollama. Use when setting up local models for development, CI pipelines, or cost reduction. Covers model selection, LangChain integration, and performance tuning.

Expert in creating Claude Code subagents, skills, slash commands, plugins, and plugin marketplaces. Automatically activates when working with .md files in .claude/ directories, agent/command/skill frontmatter, marketplace.json, or when discussing Claude Code extensibility and component creation.

This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PermissionRequest, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.

LangGraph parallel execution patterns. Use when implementing fan-out/fan-in workflows, map-reduce over tasks, or running independent agents concurrently.

Enforce mandatory npm test and npm run lint execution after every code change with no exceptions for code quality and test coverage compliance

Validate test coverage meets minimum thresholds (default 80% overall, 80% statements, 75% branches, 80% functions). Parses coverage reports from coverage/coverage-summary.json or test output. Returns pass/fail status with detailed metrics and identifies uncovered files.

LLM cost tracking with Langfuse for cached responses. Use when monitoring cache effectiveness, tracking cost savings, or attributing costs to agents in multi-agent systems.

Creates jaw-dropping, production-ready frontend interfaces AND delivers perfectly matched real photos (Unsplash/Pexels direct links) OR flawless custom image-generation prompts for hero images, backgrounds, and illustrations. Zero AI slop, zero fake URLs.

World-class prompt engineering skill for LLM optimization, prompt patterns, structured outputs, and AI product development. Expertise in Claude, GPT-4, prompt design patterns, few-shot learning, chain-of-thought, and AI evaluation. Includes RAG optimization, agent design, and LLM system architecture. Use when building AI products, optimizing LLM performance, designing agentic systems, or implementing advanced prompting techniques.

Claude Code AI-assisted development workflow. Activate when discussing Claude Code usage, AI-assisted coding, prompting strategies, or Claude Code-specific patterns.

Create PydanticAI agents with type-safe dependencies, structured outputs, and proper configuration. Use when building AI agents, creating chat systems, or integrating LLMs with Pydantic validation.

Playwright testing with autonomous test agents (planner, generator, healer)

Writing Datasette plugins using Python and the pluggy plugin system. Use when Claude needs to: (1) Create a new Datasette plugin, (2) Implement plugin hooks like prepare_connection, register_routes, render_cell, etc., (3) Add custom SQL functions, (4) Create custom output renderers, (5) Add authentication or permissions logic, (6) Extend Datasette's UI with menus, actions, or templates, (7) Package a plugin for distribution on PyPI

Execution-layer skill for spawning subagents via runAgent

Guides optimal Linear operations usage with caching, performance patterns, and error handling. Auto-activates when implementing CCPM commands that interact with Linear. Prevents usage of non-existent Linear MCP tools.