TypedDict values are rejected by type checkers for Store.put()/aput() despite being valid at runtime

[navarra-lisandro]

Checked other resources

• This is a bug, not a usage question.
• I added a clear and descriptive title that summarizes this issue.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangGraph rather than my code.
• The bug is not resolved by updating to the latest stable version of LangGraph (or the specific integration package).
• This is not related to the langchain-community package.
• I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Related Issues / PRs

No response

Reproduction Steps / Example Code (Python)

from typing import TypedDict
from langgraph.store.memory import InMemoryStore

class Prefs(TypedDict):
    theme: str

store = InMemoryStore()
value: Prefs = {"theme": "dark"}
store.put(("users", "123"), "prefs", value)

Error Message and Stack Trace (if applicable)

repro.py:9: error: Argument 3 to "put" of "BaseStore" has incompatible type "Prefs"; expected "dict[str, Any]"  [arg-type]
Found 1 error in 1 file (checked 1 source file)

Description

• I'm trying to use the langgraph library's Store.put()/aput() to store a TypedDict value (a well-typed, string-keyed mapping matching the store's own requirements).
• I expect the call to type-check cleanly with mypy, since a TypedDict is exactly the kind of typed, JSON-serializable payload the store expects.
• Instead, mypy rejects it with the error shown below, because value is typed as dict[str, Any], which requires full mutability (arbitrary keys can be added/removed/overwritten) that a TypedDict doesn't support. This breaks type-checking in CI for any caller passing a TypedDict to store.put()/aput(), despite the code working correctly at runtime.

Proposed fix: widen the value parameter on BaseStore.put()/aput() (and PutOp.value) from dict[str, Any] to collections.abc.Mapping[str, Any]. This is a compatible widening (plain dicts still satisfy Mapping[str, Any]) so it doesn't change behavior or break existing callers.

System Info

langgraph-checkpoint: 4.2.0
Python 3.14.5, Darwin