~ / search / GHSA-cc4m-mp48-x7qg

GHSA-cc4m-mp48-x7qg

Spring AI's support for Anthropic's Skills API used LLM-influenced filenames unsanitized in Path.resolve before writing files to disk

Source: GitHub Advisory Database (reviewed) Moderate CVE-2026-41863

Packages

EcosystemPackageAffectedFixed in
Maven org.springframework.ai:spring-ai-anthropic >= 1.1.0, < 1.1.7 1.1.7

Details

As written in the advisory (Markdown source, shown as plain text).

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 1.1.7

References

External links from the advisory. They lead to third-party sites.