gguf-quantization
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill directs users to clone the official llama.cpp repository from GitHub and install relevant Python packages. These resources are well-known and reputable within the AI community. \n- [COMMAND_EXECUTION]: Instructions cover standard software development workflows such as compiling C++ code with make and running Python scripts for model conversion. These operations are appropriate for the skill's technical purpose. \n- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates processing user-provided chat messages and prompts through local models. While this constitutes an injection surface, the lack of dangerous capabilities (like network exfiltration or disk modification) in the provided code snippets limits the risk to a low level. \n- [SAFE]: No evidence of obfuscated code, hidden URLs, credential theft, or persistence mechanisms was found during the analysis.
Audit Metadata