LangSmith Client SDK: Public prompt pull deserializes untrusted manifests without trust boundary warning
langchain-ai langsmith-sdk
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| langsmith PyPI |
|
|
| langsmith npm |
|
|
| langchain-classic PyPI |
|
langchain-ai langsmith-sdk
| Product | Affected | Fixed in |
|---|---|---|
| langsmith PyPI |
|
|
| langsmith npm |
|
|
| langchain-classic PyPI |
|
|
| langchain PyPI |
|
|
LangSmith Client SDKs provide SDK's for interacting with the LangSmith platform. Prior to LangSmith SDK Python 0.8.0 and JS/TS 0.6.0, the LangSmith SDK's prompt pull methods (pull_prompt / pull_prompt_commit in Python, pullPrompt / pullPromptCommit in JS/TS) fetch and deserialize prompt manifests from the LangSmith Hub. These manifests may contain serialized LangChain objects and model configuration that affect runtime behavior. When pulling a public prompt by owner/name identifier, the manifest content is controlled by an external party, but prior versions of the SDK did not distinguish this from pulling a prompt within the caller's own organization. This vulnerability is fixed in LangSmith SDK Python 0.8.0 and JS/TS 0.6.0.
From the CVE record · GitHub_M
LangSmith SDK prompt pull methods deserialize manifests that can contain LangChain objects and model configuration affecting runtime behavior. In affected versions, an external party can control that content when an application pulls a public prompt by owner/name identifier; the SDK does not distinguish it from a prompt in the caller’s own organization. Exploitation requires the application to pull such a prompt; no additional user interaction or authentication requirement is specified.
Check deployments using PyPI langsmith before 0.8.0, npm langsmith before 0.6.0, PyPI langchain-classic before 1.0.7, or PyPI langchain before 0.3.30.