What problem does it solve? Building multi-step AI agents with planning, file management, subagent delegation, and persistent memory normally requires wiring together many LangChain/LangGraph components by hand. This Skill provides the reference patterns and configuration rules for the Deep Agents framework so you get a working agent harness without implementing the middleware yourself. ## Core Features & Use Cases - Agent Creation: Create agents with create_deep_agent() using custom tools, system prompts, subagents, and model selection in Python or TypeScript. - Middleware Configuration: Set up TodoList, Filesystem, SubAgent, HumanInTheLoop, Skills, and Memory middleware with the correct backends, checkpointers, and stores. - SKILL.md Authoring: Write valid skill files with YAML frontmatter (name and description) for progressive, on-demand skill loading. - Use Case: You need an assistant that plans a multi-step research task, reads and writes files, delegates subtasks to specialized subagents, and remembers context across sessions. Use this Skill to configure the agent with a FilesystemBackend, MemorySaver checkpointer, and InMemoryStore in one create_deep_agent call. ## Quick Start Ask the agent to create a deep agent with a custom tool, a filesystem backend, a skills directory, and a checkpointer using create_deep_agent.
