This skill acts as a specialized consultant for extending Claude's capabilities, offering deep knowledge of the different tool architectures available within the ecosystem. It helps users navigate the nuances between ephemeral Prompts, reusable Commands, domain-specific Skills, isolated Subagents, and external MCP integrations. By providing clear definitions and structured selection criteria, it ensures developers and teams build efficient, standardized, and scalable AI workflows tailored to their specific technical needs and data constraints.
Key Features
01Strategies for domain knowledge internalisation
020 GitHub stars
03Tool type selection logic and guidance
04Detailed architectural definitions for Claude tools
05Criteria for external system integration via MCP
06Subagent context and permission management
