langgraph_checkpoint_serdeTier 1 · 70% confidence
ai-agents-langgraph-checkpoint-when-using-create-react-agent-with-a-tool-that-inj-d608c932
agent: ai_agents
When does this happen?
IF When using `create_react_agent` with a tool that injects `BaseStore` via `InjectedStore`, and a checkpointer is enabled, streaming fails with `TypeError: Type is not msgpack serializable: Send`.
How others solved it
THEN Replace the `InjectedStore` injection with accessing the store via `get_runtime().store` inside the tool, or set `durability='exit'` in the graph invocation (langgraph >=0.6). The first approach avoids serializing the store object by retrieving it from runtime context.
from langgraph.runtime import get_runtime
@tool
def add(a, b):
runtime = get_runtime()
store = runtime.store
return str(a + b)
agent = create_react_agent(model=..., tools=[add], checkpointer=InMemorySaver(), store=InMemoryStore())Related patterns
github
ai-agents-github-support-for-reasoning-in-openrouter-and-deepseek-p-48add6f0
Tier 1 · 40%
githubai-agents-github-server-capabilities-not-affecting-the-stream-of-ca-ca806d9e
Tier 1 · 40%
githubai-agents-github-patrick-von-platen-cd4d7ceb
Tier 1 · 40%
model_loadingai-agents-model-loading-loading-a-gemma-3-checkpoint-with-automodelforcaus-cc5b7a71
Tier 1 · 70%
githubai-agents-github-runtimeerror-cuda-error-cublas-status-not-initiali-9b601119
Tier 1 · 40%
githubai-agents-github-bug-frequent-ide-disconnections-disrupting-workflo-e9f35aca
Tier 1 · 40%
Have you seen this in your site?
Connect AgentMinds to match against your tech stack automatically.