GHSA-g28h-2cmm-rj9x: Langchain nvidia ai endpoints Langchain-nvidia-ai-endpoints Has Local File Disclosure Vulnerability
HIGHGHSA-g28h-2cmm-rj9x is a HIGH (CVSS 7.5) langchain-nvidia-ai-endpoints has local file disclosure vulnerability (CWE-22: Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')) affecting Langchain nvidia ai endpoints libraries in the PyPI ecosystem. The issue involves langchain-nvidia-ai-endpoints has local file disclosure through vlm image inputs. Remediation is available by upgrading affected packages to version 1.4.2.
GHSA-g28h-2cmm-rj9x at a Glance
Security Impact
CVSS: 7.5
The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N).
How to Fix GHSA-g28h-2cmm-rj9x
To remediate this vulnerability, upgrade all affected Langchain nvidia ai endpoints dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.
pip install langchain-nvidia-ai-endpoints>=1.4.2Run the following command in your project directory to verify that all resolved dependencies use the patched release:
pip show langchain-nvidia-ai-endpointsHow to Check if You Are Affected
To determine whether your Python environment is affected by GHSA-g28h-2cmm-rj9x, inspect your installed packages using pip:
pip show langchain-nvidia-ai-endpointsAnalyze the terminal output to locate any version of Langchain nvidia ai endpoints in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 1.4.2), your application is affected.
Note: The pip show langchain-nvidia-ai-endpoints command checks currently installed packages in your active Python environment.
Technical Details
Vulnerability Mechanism & Technical Analysis
This vulnerability is classified under CWE-22: Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal'), where the product uses external input to construct a pathname that is intended to identify a file or directory that is located underneath a restricted parent directory, but the product does not properly neutralize special elements within the pathname that can cause the pathname to resolve to a location that is outside of the restricted directory. In Langchain nvidia ai endpoints, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.
Source Advisory Statement
GHSA-g28h-2cmm-rj9x Affected Langchain nvidia ai endpoints Versions
1 component(s)| Product / Package | Ecosystem | Affected Versions | Fixed Version |
|---|---|---|---|
|
langchain-nvidia-ai-endpoints
|
PyPI |
< 1.4.2
|
1.4.2
|
References
Frequently Asked Questions
What is GHSA-g28h-2cmm-rj9x?
Which Langchain nvidia ai endpoints versions are affected?
What version fixes GHSA-g28h-2cmm-rj9x?
Is langchain-nvidia-ai-endpoints affected?
What is the CVSS score of GHSA-g28h-2cmm-rj9x?
What CWE is associated with GHSA-g28h-2cmm-rj9x?
How can I check PyPI dependencies for GHSA-g28h-2cmm-rj9x?
Machine-readable OSV JSON
OSV v1.5 Schema
{
"schema_version" : "1.4.0",
"id" : "GHSA-g28h-2cmm-rj9x",
"modified" : "2026-09-24T19:25:36Z",
"published" : "2026-09-24T19:25:36Z",
"aliases" : [ ],
"summary" : "langchain-nvidia-ai-endpoints has local file disclosure through VLM image inputs",
"details" : "## Summary\n\n`langchain-nvidia-ai-endpoints` versions before 1.4.2 accepted local filesystem paths as image inputs for Vision Language Model (VLM) requests. If an application passed attacker-controlled image input to `ChatNVIDIA` or VLM reranking APIs, an attacker could cause files readable by the application process to be read and included in the outbound request to the configured NVIDIA/NIM model endpoint.\n\n## Impact\n\nApplications are affected if they use VLM image inputs with `ChatNVIDIA` or `NVIDIARerank` and allow untrusted users to control image URLs or document image metadata.\n\nAn attacker who can control those inputs may be able to read local files accessible to the application process. Applications that do not use VLM image inputs, or that only pass trusted image references, are not affected.\n\n## Patched Version\n\nUpgrade to `langchain-nvidia-ai-endpoints >= 1.4.2`.\n\nThe patched version rejects raw local filesystem paths for VLM image inputs. Supported inputs are remote URLs, `data:image/...;base64,...` URIs, and supported asset/file IDs.\n\n## Workarounds\n\nIf upgrading is not immediately possible, applications should reject local filesystem paths in user-controlled VLM image inputs and only allow trusted remote URLs, data URIs, or known safe asset/file IDs.\n\nApplications should also run with least-privilege filesystem access so that the application process cannot read files it does not need.",
"severity" : [ {
"type" : "CVSS_V3",
"score" : "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N"
} ],
"affected" : [ {
"package" : {
"ecosystem" : "PyPI",
"name" : "langchain-nvidia-ai-endpoints"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "0"
}, {
"fixed" : "1.4.2"
} ]
} ],
"database_specific" : {
"last_known_affected_version_range" : "<= 1.4.1"
}
} ],
"references" : [ {
"type" : "WEB",
"url" : "https://github.com/langchain-ai/langchain-nvidia/security/advisories/GHSA-g28h-2cmm-rj9x"
}, {
"type" : "WEB",
"url" : "https://github.com/langchain-ai/langchain-nvidia/commit/b483c7cad07dd9d5fd0fee2668238c4a08825d86"
}, {
"type" : "PACKAGE",
"url" : "https://github.com/langchain-ai/langchain-nvidia"
}, {
"type" : "WEB",
"url" : "https://github.com/langchain-ai/langchain-nvidia/releases/tag/libs%2Fai-endpoints%2Fv1.4.2"
} ],
"database_specific" : {
"cwe_ids" : [ "CWE-22" ],
"severity" : "HIGH",
"github_reviewed" : true,
"github_reviewed_at" : "2026-09-24T19:25:36Z",
"nvd_published_at" : null
}
}