AgenticFORGE

Expert at selecting and configuring AgenticFORGE agents. Generates correct FunctionCallAgent, ReActAgent, PlanSolveAgent, ReflectionAgent, SimpleAgent, SkillAgent, and WorkflowAgent code with proper configuration. Use when the user wants to build an agent, choose between agent types, configure agent options, or understand agent behavior.

Expert at designing AgenticFORGE Skills — both Markdown SKILL.md files and TypeScript AgentSkill classes. Generates correct frontmatter, system prompts, SkillRunner setup, and SkillLoader configuration. Use when the user wants to create a Skill, define agent capabilities, set up skill routing, or load skills from a directory.

Expert at designing and implementing Tools for AgenticFORGE. Generates correct Tool subclasses, ToolChain, ToolRegistry, and AsyncToolExecutor code with proper parameter definitions and error handling. Use when the user wants to create a tool, wrap an API, chain tools, or register tools to a registry.

Expert assistant for building with AgenticFORGE SDK. Generates ready-to-run TypeScript code for agents, tools, skills, memory, RAG, and multi-agent systems. Use when the user wants to build something with AgenticFORGE, asks how to use any package, or needs working code examples.

Expert at designing and implementing Agent lifecycle hooks in AgenticFORGE. Covers hook contracts, registration patterns, strict/error behavior, event filtering, priority ordering, and production observability patterns.

Expert at using AgenticFORGE ContextBuilder to manage LLM token budgets. Use when the user needs to control token usage, prevent context overflow, or build messages that fit within a model window.

Expert at setting up AgenticFORGE memory systems — WorkingMemory, EpisodicMemory, SemanticMemory, MemoryManager, and RAGPipeline. Use when the user wants to add memory to an agent, set up RAG, persist conversation history, or store and retrieve knowledge.

Expert at diagnosing AgenticFORGE issues — tool call failures, agent loop errors, skill routing failures, memory errors, and TypeScript type errors. Use when the user reports an error, unexpected behavior, or asks how to debug AgenticFORGE code.

Expert at setting up AgenticFORGE multi-agent protocols — A2A (Agent-to-Agent) and MCP (Model Context Protocol). Use when the user wants agents to communicate, delegate tasks, expose tools via MCP, or build multi-agent pipelines.