LLM & Agents

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Core development principles and guidelines covering security, QA, performance, documentation, and coding standards. Used by all agents to ensure consistent quality across the Orchestra system.

Spec-First, Agent-Implemented Software Development Lifecycle. Use when: (1) Starting a new software project that needs structured design-before-code approach, (2) User mentions "SDLC", "spec-first", "design docs", or "implementation spec", (3) User wants to go from requirements/intent to working code with traceability, (4) Project requires documented architectural decisions and review checkpoints, (5) User has existing design artifacts and wants to continue from a specific phase. Produces: Intent doc → HLD → ADR-Lite → EIS → Code → Validation tests.

Interpretive guidance for designing Claude Code sub-agents. Helps apply official documentation effectively and avoid common pitfalls. ALWAYS use when creating or reviewing sub-agents (aka agents or subagents).

CRITICAL AUTOMATION: Automatically researches PROVEN medical mnemonics during study guide creation. ACTIVATES when creating study guides and mnemonics are mentioned. Launches the medical-mnemonic-researcher agent to find established USMLE mnemonics from medical education sources. NEVER invents mnemonics - only finds ones that already exist and are validated by medical students.

Guidelines for generating coherent book prose including voice and tone, paragraph structure, transition techniques, and chapter flow patterns. Use when writing book content, structuring chapters, or maintaining narrative consistency.

Help analyze eval logs for the Theory Ventures context agent, cluster failures by mode, and propose targeted LangGraph/LlamaIndex changes.

A test local skill for Claude. Use when testing local skills detection.

Use when creating new skills, improving existing skills, or when the user requests help with skill authoring. Creates optimized Claude Code skills (SKILL.md files) following official best practices for descriptions, structure, content patterns, and validation.

Implement Supabase database patterns for PhotoVault including RLS policies, TypeScript queries, and storage operations. Use when working with database schemas, Row Level Security, migrations, storage buckets, auth.uid() policies, or debugging empty query results. Includes PhotoVault table structure and client setup patterns.

Core orchestration skill for the Claude Organization. Manages async sub-agents, worktrees, daily logs, and context handoff. This skill defines how the main Claude Code instance operates as Secretary to the CEO.

Comprehensive systematic debugging agent for hunting down complex bugs, root causes, and error propagation across full stack (frontend, backend, APIs, databases). Use when diagnosing errors, troubleshooting issues, analyzing stack traces, investigating bugs, examining logs, debugging performance problems, or tracking down error causes. Also use when users mention debugging, errors, bugs, failures, crashes, exceptions, or unexpected behavior.

Ollama LLM inference server management via Podman Quadlet. Single-instance design with GPU acceleration for running local LLMs. Use when users need to configure Ollama, pull models, run inference, or manage the Ollama server.

Emerging technologies (prompt engineering, AI agents, red teaming) and leadership roles (product manager, engineering manager, DevRel).

Clear, concise description of what this skill does and when to use it (1-1024 chars). Include relevant keywords for agent discovery.

Orchestrate experimental procedure generation from LA-Bench format input. This skill orchestrates file I/O and delegates procedure generation logic to experiment-procedure-generator subagent. Use this skill when you need to generate step-by-step experimental procedures with quantitative specifications and proper experimental design.

AI agent skill for implementing evidence-based UI improvements. Use when tasked with UI audits, design system implementation, accessibility fixes, or frontend optimization. Provides step-by-step workflows, code patterns, and validation criteria for automated execution. Triggers on UI review, accessibility audit, design system setup, CSS refactoring, form optimization, button hierarchy, color contrast, spacing system, typography scale, or WCAG compliance tasks.

Use when creating new skills, editing existing skills, or verifying skills work before deployment - applies TDD to process documentation by testing with subagents before writing, iterating until bulletproof against rationalization

This skill should be used when the user asks about "coordinate coding agents", "orchestrate agent team", "manage multiple agents", "vibekanban workflow", "task delegation to agents", "agent swarm coordination", "parallel agent execution", "chief of staff mode", "cos mode", "you're my cos", "your my cos", "act as cos", "be my cos", "you are my chief of staff", "create tasks for agents", "dispatch agents", or needs guidance on coordinating autonomous coding agents, task breakdown strategies, or multi-agent workflow patterns.

Expert guidance for building production-ready tool-calling agents with Databricks Mosaic AI Agent Framework. Use when users need to create agents that orchestrate multiple data sources or APIs, implement LangChain-based agentic workflows, design Foundation Model-powered tool selection, optimize agent prompts and decision-making, or architect multi-tool agent systems. Covers agent architecture patterns, tool design best practices, Foundation Model integration, and common pitfalls.