LangSmith SDK: Public prompt pull deserializes untrusted manifests without trust boundary warning

Description

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.

Recommendation

Update the langsmith package to the latest compatible version. Followings are version details:

• Affected version(s): < 0.6.0
• Patched version(s): 0.6.0

References

• GHSA-3644-q5cj-c5c7
• CVE-2026-45134
• CWE-502
• CAPEC-310
• OWASP 2021-A6
• OWASP 2021-A8

Related Issues

• LangSmith Client SDK Affected by Server-Side Request Forgery via Tracing Header Injection - CVE-2026-25528
• LangSmith SDK: Streaming token events bypass output redaction - CVE-2026-41182
• LangSmith Client SDKs has Prototype Pollution in langsmith-sdk via Incomplete `__proto__` Guard in Internal lodash `set( - CVE-2026-40190
• Flowise: APIChain Prompt Injection SSRF in GET/POST API Chains - CVE-2026-41271

You might also like:

How to Secure your NodeJs Express Javascript Application - part 1

How to Secure your NodeJs Express Javascript Application - part 2

10 Secure Coding Best Practices to Follow in Every Project

Tags:

npm

langsmith