CVE-2024-5565: Vanna prompt injection leads to remote code execution via ask method

Summary

CVE-2024-5565 affects the Vanna library, which uses a prompt function to show users visualized results. Prompt injection can alter that prompt, causing arbitrary Python code to run in place of the intended visualization code. The flaw is reached through external input to the library's "ask" method when "visualize" is set to True, the default behavior, and leads to remote code execution. NVD has not yet provided an assessment.

Mitigation

The source does not state a fix yet. Check the original advisory for updates.

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

• vannaPyPILLM dependency since 2023-08-31

Related items

• MediumCVE-2026-121936: easyVoice server-side request forgery through openaiBaseUrl in /api/v1/ttsSimilar attack · NVD/CVE Database
• MediumCVE-2026-118628: MLflow get-artifacts server-side request forgery information disclosureSimilar attack · NVD/CVE Database
• HighCVE-2026-118625: MLflow Tracking Server privilege escalation via get_endpointsSimilar attack · NVD/CVE Database
• InfoWhen trust becomes the payload: Inside the threats, etc quarterly reportSimilar attack · Elastic Security Labs
• MediumCritical Langflow flaws enable authentication bypass and RCESimilar attack · CSO Online