AI & Agent Dev Bug Sandbox Back to Radar

TypeSafeClassifier Cannot Be Deserialized Due To Missing Class Mapping In Langchain-Core

TypeSafeClassifier from langchain-typesafe declares is_lc_serializable() True and dumpd() succeeds, but load() fails with both allowed_objects='all' and allowed_objects=[TypeSafeClassifier]. The root cause is that the serialized ID ('langchain', 'classifiers', 'typesafe', 'TypeSafeClassifier') is not present in langchain-core's deserialization mapping and the 'langchain_typesafe' namespace is not in the default valid namespaces.

highConfidence 92%LangChain Affected V0.0.1a3

Origin Analysis

langchain-typesafe uses a legacy namespace ID that is not registered in langchain-core's SERIALIZABLE_MAPPING, and the namespace 'langchain_typesafe' is not included in the loader's valid default namespaces. Therefore, allowed_objects='all' rejects the ID because the mapping is missing, and allowed_objects=[TypeSafeClassifier] still fails because the loader cannot import from the legacy namespace and the mapped import path is not whitelisted.

from langchain_core.load import dumpd, load from langchain_typesafe import TypeSafeClassifier api_key = 'test-key' classifier = TypeSafeClassifier(api_key=api_key) serialized = dumpd(classifier) print(serialized['id']) for allowed_objects in ('all', [TypeSafeClassifier]): try: load(serialized, allowed_objects=allowed_objects, secrets_map={'TYPESAFE_API_KEY': api_key}) except ValueError as error: print(repr(allowed_objects), error)

Fixing Code Block

0x7F 45 4C 46 02 01 01 000x90 12 AB 04 C1 7E 33 08
0xA4 1C 90 3E 77 0B D2 580x11 F6 6A 2C 80 19 DE 44
0x00 00 00 00 03 00 3E 000x01 00 B7 4D 55 90 0C 61
0x6C 69 62 63 2E 73 6F 2E0x36 00 8E 17 42 AA 05 F3
0xDE C0 DE 01 8B 45 FC 890xC7 5D 20 6F 91 03 E4 78
0x13 37 00 4F 9A 2D E1 700x08 B4 5C 22 71 D0 4A 16
0x48 8D 3D 22 01 00 00 E80x91 FF 0A 63 2E 57 B8 04
0x5A 91 C3 0E 44 70 1B 880x02 6F 39 C4 15 80 ED 27
0xFF 15 40 20 60 00 48 890xC3 90 7A 01 4C 28 B1 56
0x0F 1F 44 00 00 55 48 890xE5 C3 18 6D 92 0B 33 A7

Edge Case Audit

If langchain-typesafe is not installed, the try-import prevents errors, but the mapping will be absent and deserialization will still fail. Adding a new namespace to the whitelist could potentially allow deserialization of untrusted classes from that namespace if the package is malicious or compromised; ensure langchain-typesafe is a trusted dependency. The fix hardcodes the import path 'langchain_typesafe.classifier.TypeSafeClassifier'; if the package renames or moves the class, this mapping must be updated. Rollback: revert these changes, but note that previously serialized data will again fail to load.