langgraph-deep-agents - Build Stateful LangGraph Agent Graphs
Provides an implementation playbook for LangGraph and Deep Agents with graph control flow, checkpointing, HITL, multi-agent composition, memory, streaming, and deployment.
Tags
Updated: 2026-10-04Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Define typed graph state
- Implement conditional graph routing
- Configure checkpoint persistence
- Build human approval workflows
- Compose multi-agent systems
- Configure short-term memory
- Configure long-term memory
- Stream graph execution
- Integrate agent tools
- Deploy graph-based agents
Inputs
- Agent purpose
- State schema
- Node design
- Edge logic
- Persistence backend
- Human approval requirements
- Multi-agent pattern
- Memory requirements
- Streaming requirements
- Deployment target
- Deep Agents requirements
Outputs
- Implemented agent graph
- Checkpointed execution state
- Human interrupt and resume flow
- Configured memory stores
- Agent execution streams
- Integrated tool workflows
- Deployment configuration
- Graph visualization
- Implementation documentation
Requirements
- LangGraph runtime
- LangChain agent support
- Checkpointing backend for HITL
- Filesystem backend for Deep Agents
- LangSmith access for deployment or tracing