CVE-2026-25592: Semantic Kernel .NET SDK Arbitrary File Write in SessionsPythonPlugin
CVE-2026-25592 is an arbitrary file write, classified as CWE-22 path traversal, in the SessionsPythonPlugin of Microsoft's Semantic Kernel .NET SDK, affecting versions prior to 1.71.0. The description states the issue is fixed in Microsoft.SemanticKernel.Core 1.71.0, and the record lists a CVSS 3.1 base score of 9.9 from the GitHub CNA. No confirmed exploitation appears in the sources consulted, and the vendor describes an allow-list Function Invocation Filter as an interim mitigation.
Overview
CVE-2026-25592 affects the SessionsPythonPlugin within the Microsoft Semantic Kernel .NET SDK. According to the record, the flaw is an arbitrary file write that is classified as path traversal (CWE-22). The vendor advisory referenced in the record is GHSA-2ww3-72rp-wpp4, and the entry is also listed in the NVD record.
Impact
- An attacker able to reach the affected functionality could potentially write files to locations the application does not intend to expose.
- The GitHub CNA assigned a CVSS 3.1 base score of 9.9 (critical). The vector in the record indicates network attack vector, low attack complexity, low privileges required, no user interaction, scope changed, and high confidentiality, integrity and availability impact.
- The precise consequences depend on how an application uses the plugin and the privileges of the process running it.
Affected products and versions
- Product: Microsoft Semantic Kernel .NET SDK, component SessionsPythonPlugin.
- Per the description, versions prior to 1.71.0 are affected.
- Fixed package: Microsoft.SemanticKernel.Core version 1.71.0.
- The structured affected and fixes fields in the record are empty, so the version range is taken from the description text only.
Exploitation status
The CISA KEV catalog does not list this CVE in the data consulted. No confirmed active exploitation or exploitation in the wild appears in the sources consulted. The EPSS probability recorded is 0.02362.
Remediation and mitigation
- Upgrade Microsoft.SemanticKernel.Core to version 1.71.0 or later, which the description identifies as the fixed release.
- If upgrading is not immediately possible, the description suggests a Function Invocation Filter that checks the arguments passed to DownloadFileAsync and UploadFileAsync and restricts localFilePath to an allow list of approved locations.
- Review where the plugin is used and limit the permissions of the service account or process that runs it.
- Treat the allow-list approach as an interim control and confirm it covers every code path that reaches the affected methods.
Detection guidance
- Review application logs and telemetry for unexpected file creation or overwrites performed by components that use SessionsPythonPlugin.
- Monitor for file write operations targeting directories outside the intended working area of the application.
- Inventory .NET projects that reference Microsoft.SemanticKernel.Core and record their installed versions.
Limitations of the available information
- The record does not provide structured affected version ranges, fixed version entries or vendor-supplied CVSS details beyond the GitHub CNA score.
- The NVD analysis status in the record is Deferred, so NVD scoring and classification may change.
- The contents of the referenced advisory and pull request were not reviewed in full, so no technical detail beyond the description is asserted here.
- Whether a given deployment is exposed depends on local configuration, which is unknown from the record.
Frequently asked questions
Which versions of Semantic Kernel are affected by CVE-2026-25592?
According to the description in the record, versions prior to 1.71.0 of the Microsoft Semantic Kernel .NET SDK are affected. The structured affected version field in the record is empty, so the range should be confirmed against the vendor advisory.
Is CVE-2026-25592 known to be exploited?
The CISA KEV catalog does not list this CVE in the data consulted, and no confirmed exploitation appears in the sources consulted. The record shows an EPSS probability of 0.02362.
This analysis was drafted with AI assistance from verified source data and automatically checked against it.