Code execution via evaluated vector-store filters
Published Feb 19, 2026 · Updated Feb 19, 2026
Code injection in Microsoft Semantic Kernel before 1.39.4 allows remote authenticated users to execute code through vector-store filters. InMemoryVectorStore's default Search Plugin filter interpolates model-controlled arguments into a Python lambda, then passes it to eval after an incomplete AST blocklist. Exploitation requires a prompt-injection path and the default Search Plugin backed by the in-memory store, enabling commands with the agent host's privileges.
Summary
What happened
Code injection in Microsoft Semantic Kernel before 1.39.4 allows remote authenticated users to execute code through vector-store filters. InMemoryVectorStore's default Search Plugin filter interpolates model-controlled arguments into a Python lambda, then passes it to eval after an incomplete AST blocklist. Exploitation requires a prompt-injection path and the default Search Plugin backed by the in-memory store, enabling commands with the agent host's privileges.
The record
- CVE
- CVE-2026-26030
- Published
- Feb 19, 2026
- Updated
- Feb 19, 2026
- Vendor
- Microsoft
- Product
- Semantic Kernel
- Classifications
- CWE-94, T1059
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- Feb 19, 2026CVE publishedPublication date reported by the CVE source.
- Feb 19, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 1.39.4
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- CVE-2026-26030 AI Agent CTF challengeproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo