LangGraph Subgraphs and Composition (Python)

Overview

A parent StateGraph invokes a compiled child subgraph as a node. The child node writes state["answer"] = "42" and returns. The parent's next node reads state["answer"] and gets None. No error, no warning, no deprecation notice — just a silent None that surfaces as a wrong answer three nodes later when the router picks the "couldn't find it" branch.

The cause is pain-catalog entry P21: LangGraph subgraphs run on an independent state schema. Only keys declared in both the parent's TypedDict and the child's TypedDict propagate across the subgraph boundary. answer existed in the child schema but not the parent schema, so it was discarded on return. The fix is to declare answer in both schemas (with matching reducers, if the field is a list) or to use explicit Command(graph=ParentGraph, update={"answer": "42"}) to bubble it up.

The second silent failure waits one step further. Attach a tracing callback to the parent runnable via parent.with_config(callbacks=[tracer]) and invoke. The tracer fires on parent nodes and never on child tool calls. This is pain-catalog entry P28: LangGraph creates a fresh runtime per subgraph, so callbacks bound at definition time do not inherit. The fix is to pass callbacks at invocation time via config["callbacks"], which does propagate.

This skill walks through the shared-state contract, three dispatch patterns (compiled subgraph as a node, Send fan-out, Command(graph=Parent) bubble-up), callback scoping, per-subgraph recursion_limit budgets, and a testing pattern that exercises every subgraph in isolation before composition. Pin: langgraph 1.0.x, langchain-core 1.0.x. Pain-catalog anchors: P21, P28, with supporting references to P18 (reducers), P19 (stream modes on nested graphs), and P55 (recursion budget).

A planner-executor is typically 1 parent + 2-4 subgraphs; a hierarchical agent team with a supervisor and N specialists is 1 parent + N subgraphs. Each subgraph has its own independent recursion_limit (default 25) — a parent at step 20 can still invoke a child that runs 25 of its own steps.

Prerequisites

• Python 3.10+
• langgraph >= 1.0, < 2.0
• langchain-core >= 1.0, < 2.0
• Completion of langchain-langgraph-basics (L25) — StateGraph, TypedDict state, Annotated[list, add_messages] reducer, MemorySaver checkpointing
• Test tooling: pytest, langchain_core.language_models.fake_chat_models.FakeListChatModel

Instructions

Step 1 — Declare the shared-state contract explicitly

The single most important decision when composing subgraphs is: which keys cross the boundary? Every key that must survive the call must appear in both TypedDict schemas with compatible types and reducers.

from typing import Annotated, TypedDict
from langchain_core.messages import AnyMessage
from langgraph.graph.message import add_messages

# Keys both schemas declare -> these propagate
# Keys only in parent -> invisible to child
# Keys only in child -> discarded on return (P21)

class ParentState(TypedDict):
    # Shared with every subgraph
    messages: Annotated[list[AnyMessage], add_messages]  # P18 reducer required
    session_id: str

    # Parent-only coordination fields
    plan: list[str]
    current_step: int

class ExecutorState(TypedDict):
    # Shared with parent — must match reducer exactly (P18)
    messages: Annotated[list[AnyMessage], add_messages]
    session_id: str

    # Executor-only scratch — parent never sees these
    tool_result: dict | None
    retries: int

If messages on the child used a different reducer (or no reducer), list updates would silently replace instead of append on one side of the boundary (P18). The messages + session_id pair is the propagation contract. Everything else is private to its owner.

See State Contract for the full state-propagation matrix and the "subset rule" for schema inheritance.

Step 2 — Pick the dispatch pattern

Three ways a parent can invoke a subgraph, and each solves a different problem.

A. Compiled subgraph as a node — Simplest. Subgraph runs, returns a state update, parent continues.

from langgraph.graph import StateGraph, END

executor_graph = (
    StateGraph(ExecutorState)
    .add_node("run_tool", run_tool_node)
    .add_node("summarize", summarize_node)
    .add_edge("run_tool", "summarize")
    .add_edge("summarize", END)
    .set_entry_point("run_tool")
    .compile()
)

parent_graph = (
    StateGraph(ParentState)
    .add_node("plan", planner_node)
    .add_node("execute", executor_graph)   # compiled subgraph as a node
    .add_node("finalize", finalize_node)
    .add_edge("plan", "execute")
    .add_edge("execute", "finalize")
    .set_entry_point("plan")
    .compile()
)

Only messages and session_id cross the boundary in either direction (from Step 1). tool_result stays inside the child; plan stays inside the parent.

B. Send(graph, state) for fan-out — One parent step spawns N parallel subgraph invocations, each with a different slice of state.

from langgraph.types import Send

def dispatch_specialists(state: ParentState) -> list[Send]:
    return [
        Send("specialist_graph", {"messages": state["messages"],
                                   "session_id": state["session_id"],
                                   "topic": topic})
        for topic in state["plan"]
    ]

Use Send when the number of subgraph calls depends on runtime state. Reducers on shared keys merge the parallel results.

C. Command(graph=ParentGraph, update=...) to bubble up — A subgraph node jumps control back to the parent with an explicit state update, skipping the rest of the subgraph.

from langgraph.types import Command

def specialist_early_exit(state: ExecutorState) -> Command:
    if state.get("tool_result") and state["tool_result"].get("done"):
        return Command(
            graph=Command.PARENT,
            update={"messages": [AIMessage("done")]},
            goto="finalize",
        )
    return {"retries": state.get("retries", 0) + 1}

Command(graph=Command.PARENT) is the explicit opposite of P21 — it forces a field up to the parent scope regardless of schema overlap.

See Dispatch Patterns for the full decision tree (inline function vs subgraph-as-node vs Send vs Command vs separate service) and a sizing guide.

Step 3 — Scope callbacks at invocation time, not definition time

# WRONG — callbacks bind at definition time and do NOT propagate to subgraphs (P28)
traced_parent = parent_graph.with_config(callbacks=[tracer])
traced_parent.invoke({"messages": [HumanMessage("...")], "session_id": "s1"})
# tracer fires on parent nodes only. Child tool calls are invisible.

# RIGHT — callbacks pass via config at invocation time, propagating into every subgraph
parent_graph.invoke(
    {"messages": [HumanMessage("...")], "session_id": "s1"},
    config={
        "configurable": {"thread_id": "s1"},
        "callbacks": [tracer],
    },
)
# tracer fires on parent nodes AND every child tool, LLM, and chain event.

Every production invocation path — API handler, batch worker, test harness — should pass callbacks via config["callbacks"]. Lint for with_config(callbacks= on compiled graphs in CI and flag it.

See Callback Scoping for the debugging playbook when a callback "should be firing but isn't."

Step 4 — Budget recursion per subgraph

LangGraph's recursion_limit (default 25 supersteps) is per-graph, not global. A parent graph at superstep 20 invoking a subgraph resets the counter inside that subgraph to zero. Pros: one runaway subgraph cannot starve the parent. Cons: adding subgraphs does not reduce your global budget — a poorly bounded specialist can still rack up 25 of its own steps while the parent thinks it spent only one.

# Parent gets 10 steps of its own planning.
# Each executor call gets its own 15-step budget, independent of the parent's 10.
parent_graph.invoke(
    initial_state,
    config={
        "configurable": {"thread_id": "s1"},
        "recursion_limit": 10,
    },
)

executor_graph.invoke(
    sub_state,
    config={"recursion_limit": 15},
)

For a parent that dispatches N specialists via Send, worst-case step count is parent_limit + N * specialist_limit. Monitor the actual distribution with a callback on on_chain_start / on_chain_end at each graph boundary — GraphRecursionError with no obvious loop is P55 in the pain catalog, and subgraph composition is the most common cause.

Step 5 — Test every subgraph in isolation before composing

A subgraph that works alone and breaks in composition is almost always a state-contract bug (Step 1) or a callback-scoping bug (Step 3). Catch both by unit-testing each subgraph with a FakeListChatModel and an in-memory checkpointer before wiring it into a parent.

from langchain_core.language_models.fake_chat_models import FakeListChatModel
from langgraph.checkpoint.memory import MemorySaver

def test_executor_subgraph_standalone():
    fake = FakeListChatModel(responses=['{"done": true, "result": 42}'])
    graph = build_executor(llm=fake).compile(checkpointer=MemorySaver())
    out = graph.invoke(
        {"messages": [HumanMessage("do the thing")],
         "session_id": "test",
         "tool_result": None,
         "retries": 0},
        config={"configurable": {"thread_id": "test"},
                "recursion_limit": 5},
    )
    # Assert the shared-contract fields (Step 1) are present on return
    assert "messages" in out
    assert out["session_id"] == "test"
    # Assert child-only field is scoped correctly
    assert out["tool_result"] == {"done": True, "result": 42}

See Testing Subgraphs for the full isolation pattern — fixtures, state-shape assertions per node, and how to assert callback propagation with a capturing handler.

Step 6 — Version subgraphs independently of the parent

A reusable subgraph should ship with its own semantic version and a pinned schema contract. Breaking either the shared-state contract (Step 1) or the dispatch signature (Step 2) is a major-version bump; adding a new private child-only field is a patch.

# executor_subgraph/__init__.py
__version__ = "1.2.0"

SHARED_KEYS = frozenset({"messages", "session_id"})  # parent must declare these

def build_executor(llm) -> StateGraph:
    """v1.2.0 executor — adds 'retries' field (child-only, backward-compatible)."""
    ...

Parents pin executor_subgraph>=1.2.0,<2.0.0. A v2.0.0 that renames session_id forces every parent to re-sync its TypedDict. This is how you catch P21 at pip install time instead of at runtime.

Output

• ParentState + per-subgraph child TypedDicts with the shared-key contract declared explicitly; every shared list field has a matching reducer
• Dispatch pattern chosen per subgraph (node vs Send vs Command(graph=PARENT)) with rationale recorded in code comments
• Every invocation path (API, batch, test) passes callbacks via config["callbacks"] so observability propagates into every subgraph
• recursion_limit set explicitly on parent and each subgraph, with the worst-case superstep count documented
• Unit tests for each subgraph in isolation, asserting shared-contract fields on return and callback propagation via a capturing handler
• Reusable subgraphs ship with their own __version__ and a frozen SHARED_KEYS set

Error Handling

Examples

Planner-executor — shared messages, private scratch

The prototypical 1 + 2-4 composition: a planner writes a step list, an executor runs each step, a finalizer summarizes. messages and session_id cross every boundary; plan is parent-only; tool_result and retries are executor-only. Uses a compiled-subgraph-as-node dispatch (Step 2A). See Dispatch Patterns for the full worked example.

Hierarchical agent team — supervisor fans out to N specialists

A supervisor picks which specialists to invoke and calls Send("specialist", ...) in parallel. Each specialist runs its own subgraph with its own recursion budget. Results merge via the messages reducer. See Dispatch Patterns for the Send-based fan-out and the callback propagation pattern needed to trace the team.

Reusable subgraph library — pinned contract with versioned exports

An executor_subgraph package exports build_executor(llm) and a frozen SHARED_KEYS set, ships with a semantic version, and its CI fails if SHARED_KEYS changes without a major-version bump. See State Contract for the schema-pinning pattern.

Testing a specialist with a capturing callback

A CaptureHandler subclasses BaseCallbackHandler, appends every on_* event to a list, and asserts both parent and child events are present after a single parent_graph.invoke(..., config={"callbacks": [handler]}). Catches P28 at PR review. See Testing Subgraphs for the full fixture and assertion pattern.

Resources

• LangGraph subgraph concepts
• LangGraph subgraph how-to
• Send API reference
• Command API reference
• LangGraph callback propagation
• LangChain 1.0 release notes
• Sibling skill: langchain-langgraph-basics (L25) — prerequisite StateGraph, reducers, checkpointing
• Pack pain catalog: docs/pain-catalog.md (entries P18, P19, P21, P28, P55)