Skip to content

CVE-2026-41863 LLM-influenced filename used unsanitized in Path.resolve before file write in Spring AI support for Anthropic Skills API

MediumCVSS 6.5 EPSS 0.40%
Published
Updated
Exploit status
No public exploit known
CISA SSVC
No known exploitation · automatable: no · technical impact: partial

Description

Spring AI's support for Anthropic's Skills API used LLM-influenced filenames unsanitized in Path.resolve before writing files to disk. This could allow a malicious user to write files outside the intended target directory, including restricted directories.

Affected versions:
Spring AI: 1.1.0 through 1.1.x

Upgrade path

3 packages: 3 fixed.

PackageAffectedFixed inStatus
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.0; = 1.1.1; = 1.1.2; = 1.1.3 +4 more 1.1.7 Fixed
Spring AI >= 1.1.0, <= 1.1.x after 1.1.x Fixed
VMware Spring AI >= 1.1.0, < 1.1.7 1.1.7 Fixed

Affected products

ProductAffected versionsFixed in
Spring AI >= 1.1.0, <= 1.1.x No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) >= 1.1.0, < 1.1.7 1.1.7
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.0 No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.1 No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.2 No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.3 No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.4 No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.5 No fix listed
org.springframework.ai/spring-ai-anthropic(Maven) = 1.1.6 No fix listed
VMware Spring AI >= 1.1.0, < 1.1.7 1.1.7

References