langgraph · bobmatnyc — Claude Code skill: what it does, how to install

---
name: langgraph
description: LangGraph framework for building stateful, multi-agent AI applications with cyclical workflows, human-in-the-loop patterns, and persistent checkpointing.
user-invocable: false
disable-model-invocation: true
progressive_disclosure:
  entry_point:
    - summary
    - when_to_use
    - quick_start
  sections:
    - core_concepts
    - graph_construction
    - state_management
    - conditional_routing
    - multi_agent_patterns
    - human_in_loop
    - checkpointing_persistence
    - time_travel_debugging
    - streaming
    - tool_integration
    - error_handling
    - subgraphs_composition
    - map_reduce
    - production_deployment
    - langsmith_integration
    - real_world_examples
    - performance_optimization
    - comparison_simple_chains
    - migration_langchain
    - best_practices
estimated_tokens:
  entry: 80
  full: 5800
---

# LangGraph Workflows

## Summary

LangGraph is a framework for building **stateful, multi-agent applications** with LLMs. It implements state machines and directed graphs for orchestration, enabling complex workflows with persistent state management, human-in-the-loop support, and time-travel debugging.

**Key Innovation**: Transforms agent coordination from sequential chains into cyclic graphs with persistent state, conditional branching, and production-grade debugging capabilities.

## When to Use

✅ **Use LangGraph When**:
- Multi-agent coordination required
- Complex state management needs
- Human-in-the-loop workflows (approval gates, reviews)
- Need debugging/observability (time-travel, replay)
- Conditional branching based on outputs
- Building production agent systems
- State persistence across sessions

❌ **Don't Use LangGraph When**:
- Simple single-agent tasks
- No state persistence needed
- Prototyping/experimentation phase (use simple chains)
- Team lacks graph/state machine expertise
- Stateless request-response patterns

## Quick Start

```python
from langgraph.graph import StateGraph, END
from typing import TypedDict, Annotated
import operator

# 1. Define state schema
class AgentState(TypedDict):
    messages: Annotated[list, operator.add]
    current_step: str

# 2. Create graph
workflow = StateGraph(AgentState)

# 3. Add nodes (agents)
def researcher(state):
    return {"messages": ["Research complete"], "current_step": "research"}

def writer(state):
    return {"messages": ["Article written"], "current_step": "writing"}

workflow.add_node("researcher", researcher)
workflow.add_node("writer", writer)

# 4. Add edges (transitions)
workflow.add_edge("researcher", "writer")
workflow.add_edge("writer", END)

# 5. Set entry point and compile
workflow.set_entry_point("researcher")
app = workflow.compile()

# 6. Execute
result = app.invoke({"messages": [], "current_step": "start"})
print(result)
```

---

## Core Concepts

### StateGraph

The fundamental building block representing a directed graph of agents with shared state.

```python
from langgraph.graph import StateGraph, END
from typing import TypedDict

class AgentState(TypedDict):
    """State schema shared across all nodes."""
    messages: list
    user_input: str
    final_output: str
    metadata: dict

# Create graph with state schema
workflow = StateGraph(AgentState)
```

**Key Properties**:
- **Nodes**: Agent functions that transform state
- **Edges**: Transitions between nodes (static or conditional)
- **State**: Shared data structure passed between nodes
- **Entry Point**: Starting node of execution
- **END**: Terminal node signaling completion

### Nodes

Nodes are functions that receive current state and return state updates.

```python
def research_agent(state: AgentState) -> dict:
    """Node function: receives state, returns updates."""
    query = state["user_input"]

    # Perform research (simplified)
    results = search_web(query)

    # Return state updates (partial state)
    return {
        "messages": state["messages"] + [f"Research: {results}"],
        "metadata": {"research_complete": True}
    }

# Add node to graph
workflow.add_node("researcher", research_agent)
```

**Node Behavior**:
- Receives **full state** as input
- Returns **partial state** (only fields to update)
- Can be sync or async functions
- Can invoke LLMs, call APIs, run computations

### Edges

Edges define transitions between nodes.

#### Static Edges

```python
# Direct transition: researcher → writer
workflow.add_edge("researcher", "writer")

# Transition to END
workflow.add_edge("writer", END)
```

#### Conditional Edges

```python
def should_continue(state: AgentState) -> str:
    """Routing function: decides next node based on state."""
    last_message = state["messages"][-1]

    if "APPROVED" in last_message:
        return END
    elif "NEEDS_REVISION" in last_message:
        return "writer"
    else:
        return "reviewer"

workflow.add_conditional_edges(
    "reviewer",  # Source node
    should_continue,  # Routing function
    {
        END: END,
        "writer": "writer",
        "reviewer": "reviewer"
    }
)
```

---

## Graph Construction

### Complete Workflow Example

```python
from langgraph.graph import StateGraph, END
from typing import TypedDict, Annotated
import operator
from langchain_anthropic import ChatAnthropic

# State schema with reducer
class ResearchState(TypedDict):
    topic: str
    research_notes: Annotated[list, operator.add]  # Reducer: appends to list
    draft: str
    revision_count: int
    approved: bool

# Initialize LLM
llm = ChatAnthropic(model="claude-sonnet-4")

# Node 1: Research
def research_node(state: ResearchState) -> dict:
    """Research the topic and gather information."""
    topic = state["topic"]

    prompt = f"Research key points about: {topic}"
    response = llm.invoke(prompt)

    return {
        "research_notes": [response.content]
    }

# Node 2: Write
def write_node(state: ResearchState) -> dict:
    """Write draft based on research."""
    notes = "\n".join(state["research_notes"])

    prompt = f"Write article based on:\n{notes}"
    response = llm.invoke(prompt)

    return {
        "draft": response.content,
        "revision_count": state.get("revision_count", 0)
    }

# Node 3: Review
def review_node(state: ResearchState) -> dict:
    """Review the draft and decide if approved."""
    draft = state["draft"]

    prompt = f"Review this draft. Reply APPROVED or NEEDS_REVISION:\n{draft}"
    response = llm.invoke(prompt)

    approved = "APPROVED" in response.content

    return {
        "approved": approved,
        "revision_count": state["revision_count"] + 1,
        "research_notes": [f"Review feedback: {response.content}"]
    }

# Routing logic
def should_continue_writing(state: ResearchState) -> str:
    """Decide next step after review."""
    if state["approved"]:
        return END
    elif state["revision_count"] >= 3:
        return END  # Max revisions reached
    else:
        return "writer"

# Build graph
workflow = StateGraph(ResearchState)

workflow.add_node("researcher", research_node)
workflow.add_node("writer", write_node)
workflow.add_node("reviewer", review_node)

# Static edges
workflow.add_edge("researcher", "writer")
workflow.add_edge("writer", "reviewer")

# Conditional edge from reviewer
workflow.add_conditional_edges(
    "reviewer",
    should_continue_writing,
    {
        END: END,
        "writer": "writer"
    }
)

workflow.set_entry_point("researcher")

# Compile
app = workflow.compile()

# Execute
result = app.invoke({
    "topic": "AI Safety",
    "research_notes": [],
    "draft": "",
    "revision_count": 0,
    "approved": False
})

print(f"Final draft: {result['draft']}")
print(f"Revisions: {result['revision_count']}")
```

---

## State Management

### State Schema with TypedDict

```python
from typing import TypedDict, Annotated, Literal
import operator

class WorkflowState(TypedDict):
    # Simple fields (replaced on update)
    user_id: str
    request_id: str
    status: Literal["pending", "processing", "complete", "failed"]

    # List with reducer (appends instead of replacing)
    messages: Annotated[list, operator.add]

    # Dict with custom reducer
    metadata: Annotated[dict, lambda x, y: {**x, **y}]

    # Optional fields
    error: str | None
    result: dict | None
```

### State Reducers

Reducers control how state updates are merged.

```python
import operator
from typing import Annotated

# Built-in reducers
Annotated[list, operator.add]  # Append to list
Annotated[set, operator.or_]   # Union of sets
Annotated[int, operator.add]   # Sum integers

# Custom reducer
def merge_dicts(existing: dict, update: dict) -> dict:
    """Deep merge dictionaries."""
    result = existing.copy()
    for key, value in update.items():
        if key in result and isinstance(result[key], dict) and isinstance(value, dict):
            result[key] = merge_dicts(result[key], value)
        else:
            result[key] = value
    return result

class State(TypedDict):
    config: Annotated[dict, merge_dicts]
```

### State Updates

Nodes return **partial state** - only fields to update:

```python
def node_function(state: State) -> dict:
    """Return only fields that should be updated."""
    return {
        "messages": ["New message"],  # Will be appended
        "status": "processing"  # Will be replaced
    }
    # Other fields remain unchanged
```

---

## Conditional Routing

### Basic Routing Function

```python
def route_based_on_score(state: State) -> str:
    """Route to different nodes based on state."""
    score = state["quality_score"]

    if score >= 0.9:
        return "publish"
    elif score >= 0.6:
        return "review"
    else:
        return "revise"

workflow.add_conditional_edges(
    "evaluator",
    route_based_on_score,
    {
        "publish": "publisher",
        "review": "reviewer",
        "revise": "reviser"
    }
)
```

### Dynamic Routing with LLM

```python
from langchain_anthropic import ChatAnthropic

def llm_router(state: State) -> str:
    """Use LLM to decide next step."""
    llm = ChatAnthropic(model="claude-sonnet-4")

    prompt = f"""
    Current state: {state['current_step']}
    User request: {state['user_input']}

    Which agent should handle this next?
    Options: researcher, coder, writer, FINISH

    Return only one word.
    """

    response = llm.invoke(prompt)
    next_agent = response.content.strip().lower()

    if next_agent == "finish":
        return END
    else:
        return next_agent

workflow.add_conditional_edges(
    "supervisor",
    llm_router,
    {
        "researcher": "researcher",
        "coder": "coder",
        "writer": "writer",
        END: END
    }
)
```

### Multi-Condition Routing

```python
def complex_router(state: State) -> str:
    """Route based on multiple conditions."""
    # Check multiple conditions
    has_errors = bool(state.get("errors"))
    is_approved = state.get("approved", False)
    iteration_count = state.get("iterations", 0)

    # Priority-based routing
    if has_errors:
        return "error_handler"
    elif is_approved:
        return END
    elif iteration_count >= 5:
        return "escalation"
    else:
        return "processor"

workflow.add_conditional_edges(
    "validator",
    complex_router,
    {
        "error_handler": "error_handler",
        "processor": "processor",
        "escalation": "escalation",
        END: END
    }
)
```

---

## Multi-Agent Patterns

### Supervisor Pattern

One supervisor coordinates multiple specialized agents.

```python
from langgraph.graph import StateGraph, END
from langchain_anthropic import ChatAnthropic
from typing import TypedDict, Literal

class SupervisorState(TypedDict):
    task: str
    agent_history: list
    result: dict
    next_agent: str

# Specialized agents
class ResearchAgent:
    def __init__(self):
        self.llm = ChatAnthropic(model="claude-sonnet-4")

    def run(self, state: SupervisorState) -> dict:
        response = self.llm.invoke(f"Research: {state['task']}")
        return {
            "agent_history": [f"Research: {response.content}"],
            "result": {"research": response.content}
        }

class CodingAgent:
    def __init__(self):
        self.llm = ChatAnthropic(model="claude-sonnet-4")

    def run(self, state: SupervisorState) -> dict:
        response = self.llm.invoke(f"Code: {state['task']}")
        return {
            "agent_history": [f"Code: {response.content}"],
            "result": {"code": response.content}
        }

class SupervisorAgent:
    def __init__(self):
        self.llm = ChatAnthropic(model="claude-sonnet-4")

    def route(self, state: SupervisorState) -> dict:
        """Decide which agent to use next."""
        history = "\n".join(state.get("agent_history", []))

        prompt = f"""
        Task: {state['task']}
        Progress: {history}

        Which agent should handle the next step?
        Options: researcher, coder, FINISH

        Return only one word.
        """

        response = self.llm.invoke(prompt)
        next_agent = response.content.strip().lower()

        return {"next_agent": next_agent}

# Build supervisor workflow
def create_supervisor_workflow():
    workflow = StateGraph(SupervisorState)

    # Initialize agents
    research_agent = ResearchAgent()
    coding_agent = CodingAgent()
    supervisor = SupervisorAgent()

    # Add nodes
    workflow.add_node("supervisor", supervisor.route)
    workflow.add_node("researcher", research_agent.run)
    workflow.add_node("coder", coding_agent.run)

    # Conditional routing from supervisor
    def route_from_supervisor(state: SupervisorState) -> str:
        next_agent = state.get("next_agent", "FINISH")
        if next_agent == "finish":
            return END
        return next_agent

    workflow.add_conditional_edges(
        "supervisor",
        route_from_supervisor,
        {
            "researcher": "researcher",
            "coder": "coder",
            END: END
        }
    )

    # Loop back to supervisor after each agent
    workflow.add_edge("researcher", "supervisor")
    workflow.add_edge("coder", "supervisor")

    workflow.set_entry_point("supervisor")

    return workflow.compile()

# Execute
app = create_supervisor_workflow()
result = app.invoke({
    "task": "Build a REST API for user management",
    "agent_history": [],
    "result": {},
    "next_agent": ""
})
```

### Hierarchical Multi-Agent

Nested supervisor pattern with sub-teams.

```python
class TeamState(TypedDict):
    task: str
    team_results: dict

def create_backend_team():
    """Sub-graph for backend development."""
    workflow = StateGraph(TeamState)

    workflow.add_node("api_designer", design_api)
    workflow.add_node("database_designer", design_db)
    workflow.add_node("implementer", implement_backend)

    workflow.add_edge("api_designer", "database_designer")
    workflow.add_edge("database_designer", "implementer")
    workflow.add_edge("implementer", END)

    workflow.set_entry_point("api_designer")

    return workflow.compile()

def create_frontend_team():
    """Sub-graph for frontend development."""
    workflow = StateGraph(TeamState)

    workflow.add_node("ui_designer", design_ui)
    workflow.add_node("component_builder", build_components)
    workflow.add_node("integrator", integrate_frontend)

    workflow.add_edge("ui_designer", "component_builder")
    workflow.add_edge("component_builder", "integrator")
    workflow.add_edge("integrator", END)

    workflow.set_entry_point("ui_designer")

    return workflow.compile()

# Top-level coordinator
def create_project_workflow():
    workflow = StateGraph(TeamState)

    # Add team sub-graphs as nodes
    backend_team = create_backend_team()
    frontend_team = create_frontend_team()

    workflow.add_node("backend_team", backend_team)
    workflow.add_node("frontend_team", frontend_team)
    workflow.add_node("integrator", integrate_teams)

    # Parallel execution of teams
    workflow.add_edge("backend_team", "integrator")
    workflow.add_edge("frontend_team", "integrator")
    workflow.add_edge("integrator", END)

    workflow.set_entry_point("backend_team")

    return workflow.compile()
```

### Swarm Pattern (2025)

Dynamic agent hand-offs with collaborative decision-making.

```python
from typing import Literal

class SwarmState(TypedDict):
    task: str
    messages: list
    current_agent: str
    handoff_reason: str

def create_swarm():
    """Agents can dynamically hand off to each other."""

    def research_agent(state: SwarmState) -> dict:
        llm = ChatAnthropic(model="claude-sonnet-4")

        # Perform research
        response = llm.invoke(f"Research: {state['task']}")

        # Decide if handoff needed
        needs_code = "implementation" in response.content.lower()

        if needs_code:
            return {
                "messages": [response.content],
                "current_agent": "coder",
                "handoff_reason": "Need implementation"
            }
        else:
            return {
                "messages": [response.content],
                "current_agent": "writer",
                "handoff_reason": "Ready for documentation"
            }

    def coding_agent(state: SwarmState) -> dict:
        llm = ChatAnthropic(model="claude-sonnet-4")

        response = llm.invoke(f"Implement: {state['task']}")

        return {
            "messages": [response.content],
            "current_agent": "tester",
            "handoff_reason": "Code complete, needs testing"
        }

    def tester_agent(state: SwarmState) -> dict:
        # Test the code
        tests_pass = True  # Simplified

        if tests_pass:
            return {
                "messages": ["Tests passed"],
                "current_agent": "DONE",
                "handoff_reason": "All tests passing"
            }
        else:
            return {
                "messages": ["Tests failed"],
                "current_agent": "coder",
                "handoff_reason": "Fix failing tests"
            }

    # Build graph
    workflow = StateGraph(SwarmState)

    workflow.add_node("researcher", research_agent)
    workflow.add_node("coder", coding_agent)
    workflow.add_node("tester", tester_agent)

    # Dynamic routing based on current_agent
    def route_swarm(state: SwarmState) -> str:
        next_agent = state.get("current_agent", "DONE")
        if next_agent == "DONE":
            return END
        return next_agent

    workflow.add_conditional_edges(
        "researcher",
        route_swarm,
        {"coder": "coder", "writer": "writer"}
    )

    workflow.add_conditional_edges(
        "coder",
        route_swarm,
        {"tester": "tester"}
    )

    workflow.add_conditional_edges(
        "tester",
        route_swarm,
        {"coder": "coder", END: END}
    )

    workflow.set_entry_point("researcher")

    return workflow.compile()
```

---

## Human-in-the-Loop

### Interrupt for Approval

```python
from langgraph.checkpoint.sqlite import SqliteSaver
from langgraph.graph import StateGraph, END

# Enable checkpointing for interrupts
memory = SqliteSaver.from_conn_string(":memory:")

def create_approval_workflow():
    workflow = StateGraph(dict)

    workflow.add_node("draft", create_draft)
    workflow.add_node("approve", human_approval)  # Interrupt point
    workflow.add_node("publish", publish_content)

    workflow.add_edge("draft", "approve")
    workflow.add_edge("approve", "publish")
    workflow.add_edge("publish", END)

    workflow.set_entry_point("draft")

    # Compile with interrupt before "approve"
    return workflow.compile(
        checkpointer=memory,
        interrupt_before=["approve"]  # Pause here
    )

# Usage
app = create_approval_workflow()
config = {"configurable": {"thread_id": "1"}}

# Step 1: Run until interrupt
for event in app.stream({"content": "Draft article..."}, config):
    print(event)
    # Workflow pauses at "approve" node

# Human reviews draft
print("Draft ready for review. Approve? (y/n)")
approval = input()

# Step 2: Resume with approval decision
if approval == "y":
    # Continue execution
    result = app.invoke(None, config)  # Resume from checkpoint
    print("Published:", result)
else:
    # Optionally update state and retry
    app.update_state(config, {"needs_revision": True})
    result = app.invoke(None, config)
```

### Interactive Approval Gates

```python
class ApprovalState(TypedDict):
    draft: str
    approved: bool
    feedback: str

def approval_node(state: ApprovalState) -> dict:
    """This node requires human interaction."""
    # This is where workflow pauses
    # Human provides input through update_state
    return state  # No automatic changes

def create_interactive_workflow():
    workflow = StateGraph(ApprovalState)

    workflow.add_node("writer", write_draft)
    workflow.add_node("approval_gate", approval_node)
    workflow.add_node("reviser", revise_draft)
    workflow.add_node("publisher", publish_final)

    workflow.add_edge("writer", "approval_gate")

    # Route based on approval
    def route_after_approval(state: ApprovalState) -> str:
        if state.get("approved"):
            return "publisher"
        else:
            return "reviser"

    workflow.add_conditional_edges(
        "approval_gate",
        route_after_approval,
        {"publisher": "publisher", "reviser": "reviser"}
    )

    workflow.add_edge("reviser", "approval_gate")  # Loop back
    workflow.add_edge("publisher", END)

    workflow.set_entry_point("writer")

    memory = SqliteSaver.from_conn_string("approvals.db")
    return workflow.compile(
        checkpointer=memory,
        interrupt_before=["approval_gate"]
    )

# Execute with human intervention
app = create_interactive_workflow()
config = {"configurable": {"thread_id": "article_123"}}

# Step 1: Generate draft (pauses at approval_gate)
for event in app.stream({"draft": "", "approved": False, "feedback": ""}, config):
    print(event)

# Step 2: Human reviews and provides feedback
current_state = app.get_state(config)
print(f"Review this draft: {current_state.values['draft']}")

# Human decides
app.update_state(config, {
    "approved": False,
    "feedback": "Add more examples in section 2"
})

# Step 3: Resume (goes to reviser due to approved=False)
result = app.invoke(None, config)
```

---

## Checkpointing and Persistence

### MemorySaver (In-Memory)

```python
from langgraph.checkpoint.memory import MemorySaver

# In-memory checkpointing (lost on restart)
memory = MemorySaver()

app = workflow.compile(checkpointer=memory)

config = {"configurable": {"thread_id": "conversation_1"}}

# State automatically saved at each step
result = app.invoke(initial_state, config)
```

### SQLite Persistence

```python
from langgraph.checkpoint.sqlite import SqliteSaver

# Persistent checkpointing with SQLite
checkpointer = SqliteSaver.from_conn_string("./workflow_state.db")

app = workflow.compile(checkpointer=checkpointer)

config = {"configurable": {"thread_id": "user_session_456"}}

# State persists across restarts
result = app.invoke(initial_state, config)

# Later: Resume from same thread
result2 = app.invoke(None, config)  # Continues from last state
```

### PostgreSQL for Production

```python
from langgraph.checkpoint.postgres import PostgresSaver

# Production-grade persistence
checkpointer = PostgresSaver.from_conn_string(
    "postgresql://user:pass@localhost/langgraph_db"
)

app = workflow.compile(checkpointer=checkpointer)

# Supports concurrent workflows with isolation
config1 = {"configurable": {"thread_id": "user_1"}}
config2 = {"configurable": {"thread_id": "user_2"}}

# Each thread maintains independent state
result1 = app.invoke(state1, config1)
result2 = app.invoke(state2, config2)
```

### Redis for Distributed Systems

```python
from langgraph.checkpoint.redis import RedisSaver

# Distributed checkpointing
checkpointer = RedisSaver.from_conn_string(
    "redis://localhost:6379",
    ttl=3600  # 1 hour TTL
)

app = workflow.compile(checkpointer=checkpointer)

# Supports horizontal scaling
# Multiple workers can process different threads
```

---

## Time-Travel Debugging

### View State History

```python
from langgraph.checkpoint.sqlite import SqliteSaver

checkpointer = SqliteSaver.from_conn_string("workflow.db")
app = workflow.compile(checkpointer=checkpointer)

config = {"configurable": {"thread_id": "debug_session"}}

# Run workflow
result = app.invoke(initial_state, config)

# Retrieve full history
history = app.get_state_history(config)

for i, state_snapshot in enumerate(history):
    print(f"Step {i}:")
    print(f"  Node: {state_snapshot.next}")
    print(f"  State: {state_snapshot.values}")
    print(f"  Metadata: {state_snapshot.metadata}")
    print()
```

### Replay from Checkpoint

```python
# Get state at specific step
history = list(app.get_state_history(config))
checkpoint_3 = history[3]  # Get state at step 3

# Replay from that checkpoint
config_replay = {
    "configurable": {
        "thread_id": "debug_session",
        "checkpoint_id": checkpoint_3.config["configurable"]["checkpoint_id"]
    }
}

# Resume from step 3
result = app.invoke(None, config_replay)
```

### Rewind and Modify

```python
# Rewind to step 5
history = list(app.get_state_history(config))
step_5 = history[5]

# Update state at that point
app.update_state(
    config,
    {"messages": ["Modified message"]},
    as_node="researcher"  # Apply update as if from this node
)

# Continue execution with modified state
result = app.invoke(None, config)
```

### Debug Workflow

```python
def debug_workflow():
    """Debug workflow step-by-step."""
    checkpointer = SqliteSaver.from_conn_string("debug.db")
    app = workflow.compile(checkpointer=checkpointer)

    config = {"configurable": {"thread_id": "debug"}}

    # Execute step-by-step
    state = initial_state
    for step in range(10):  # Max 10 steps
        print(f"\n=== Step {step} ===")

        # Get current state
        current = app.get_state(config)
        print(f"Current state: {current.values}")
        print(f"Next node: {current.next}")

        if not current.next:
            print("Workflow complete")
            break

        # Execute one step
        for event in app.stream(None, config):
            print(f"Event: {event}")

        # Pause for inspection
        input("Press Enter to continue...")

    # View full history
    print("\n=== Full History ===")
    for i, snapshot in enumerate(app.get_state_history(config)):
        print(f"Step {i}: {snapshot.next} -> {snapshot.values}")

debug_workflow()
```

---

## Streaming

### Token Streaming

```python
from langchain_anthropic import ChatAnthropic

def streaming_node(state: dict) -> dict:
    """Node that streams LLM output."""
    llm = ChatAnthropic(model="claude-sonnet-4")

    # Use streaming
    response = ""
    for chunk in llm.stream(state["prompt"]):
        content = chunk.content
        print(content, end="", flush=True)
        response += content

    return {"response": response}

# Stream events from workflow
for event in app.stream(initial_state):
    print(event)
```

### Event Streaming

```python
# Stream all events (node execution, state updates)
for event in app.stream(initial_state, stream_mode="updates"):
    node_name = event.keys()[0]
    state_update = event[node_name]
    print(f"Node '{node_name}' updated state: {state_update}")
```

### Custom Streaming

```python
from typing import AsyncGenerator

async def streaming_workflow(input_data: dict) -> AsyncGenerator:
    """Stream workflow progress in real-time."""
    config = {"configurable": {"thread_id": "stream"}}

    async for event in app.astream(input_data, config):
        # Stream each state update
        yield {
            "type": "state_update",
            "data": event
        }

    # Stream final result
    final_state = app.get_state(config)
    yield {
        "type": "complete",
        "data": final_state.values
    }

# Usage in FastAPI
from fastapi import FastAPI
from fastapi.responses import StreamingResponse

app_api = FastAPI()

@app_api.post("/workflow/stream")
async def stream_workflow(input: dict):
    return StreamingResponse(
        streaming_workflow(input),
        media_type="text/event-stream"
    )
```

---

## Tool Integration

### LangChain Tools

```python
from langchain.tools import Tool
from langchain_community.tools import DuckDuckGoSearchRun
from langchain.agents import AgentExecutor, create_react_agent

# Define tools
search = DuckDuckGoSearchRun()

tools = [
    Tool(
        name="Search",
        func=search.run,
        description="Search the web for information"
    )
]

def agent_with_tools(state: dict) -> dict:
    """Node that uses LangChain tools."""
    from langchain_anthropic import ChatAnthropic

    llm = ChatAnthropic(model="claude-sonnet-4")

    # Create agent with tools
    agent = create_react_agent(llm, tools, prompt_template)
    agent_executor = AgentExecutor(agent=agent, tools=tools)

    # Execute with tools
    result = agent_executor.invoke({"input": state["query"]})

    return {"result": result["output"]}

# Add to graph
workflow.add_node("agent_with_tools", agent_with_tools)
```

### Custom Tools

```python
from langchain.tools import StructuredTool
from pydantic import BaseModel, Field

class CalculatorInput(BaseModel):
    expression: str = Field(description="Mathematical expression to evaluate")

def calculate(expression: str) -> str:
    """Evaluate mathematical expression."""
    try:
        result = eval(expression)  # Simplified
        return f"Result: {result}"
    except Exception as e:
        return f"Error: {e}"

calculator_tool = StructuredTool.from_function(
    func=calculate,
    name="Calculator",
    description="Evaluate mathematical expressions",
    args_schema=CalculatorInput
)

tools = [calculator_tool]

def tool_using_node(state: d...