The Agent Skills Directory

Standard Framework Documentation: The skill provides accurate and safe guidance on using the LangGraph library for both Python and TypeScript. It covers essential concepts like state management, nodes, edges, and error handling without introducing any malicious patterns.
Secure State Management: The instructions emphasize returning partial updates and using reducers for state modification. This is a secure approach that prevents unintended state mutation and ensures data integrity within the graph.
Input Handling Best Practices: Code examples demonstrate using structured schemas (TypedDict in Python, Zod in TypeScript) to define state, which encourages validation of data as it moves through the agent workflow.
Legitimate Library Usage: All referenced packages and imports (e.g., @langchain/langgraph, langgraph, zod) are standard, well-known libraries from a reputable source, used here for their intended purpose.

langgraph-fundamentals