llama-cpp

Warn

Audited by Snyk on Mar 28, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 0.80). The SKILL.md "Download model" section explicitly instructs fetching models from public third-party sites (e.g., huggingface-cli download TheBloke/Llama-2-7B-Chat-GGUF and links to https://huggingface.co and GitHub/Discord), which are untrusted user-hosted artifacts that the agent will load and that can materially influence its outputs and actions.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 28, 2026, 06:08 PM
Issues
1