AXME

Add durable execution and human-in-the-loop to LangGraph in 10 minutes

Integrate AXME Cloud with LangGraph.

LangGraph defines how your agent thinks. AXME wraps graph execution in durable intents — so approvals, tool waits, and handoffs survive restarts and deploys.

Quickstart (5 steps)

# 1. pip install axme langgraph
# 2. wrap graph compile with axme.adapter
# 3. submit intent before kickoff
# 4. await wait_for_human at approval nodes
# 5. complete intent when graph finishes

Approval node with AXME

LangGraph

def approve_node(state):
    return {"approved": True}

LangGraph + AXME

await axme.wait_for_human(
  task="approve", assignee=manager
)

PATTERNS

Common production patterns.

Human approval gate

Pause graph until manager responds.

Long tool wait

WAITING_FOR_TOOL for external APIs.

Multi-graph handoff

Shared intent across graphs.

Related capabilities

AXME vs LangGraph

Complementary layers.

Human-in-the-loop

Eight HITL types.

Durable waiting

Four wait states.

Ship your first durable agent — in under 10 minutes.

Free tier. No credit card. Self-host or hosted — your choice.

Start free now Read the docs