This skill is for designing, generating MVP blueprints, auditing, refactoring, or explaining agentic harnesses across domains. It covers provider-neutral agent architecture for OpenAI, Anthropic, and OpenAI-compatible APIs, including agent loops, tool design, permissions, system prompts, planning, goals, context compaction, memory, skills, MCP/external connectors, observability, evals, and prompt caching.
Desenvolvimento#mcp#aiby DenisSergeevitch
