huggingface-local-models
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate instructions for setting up and using llama.cpp. All external references and downloads target well-known and trusted platforms.
- [COMMAND_EXECUTION]: The skill contains commands to clone and compile the official llama.cpp repository from GitHub (
github.com/ggml-org/llama.cpp). These are standard procedures for the tool's intended use and do not involve suspicious execution patterns. - [EXTERNAL_DOWNLOADS]: The skill interacts with Hugging Face (
huggingface.co) to search for and download model files. These operations use official Hugging Face APIs and tools, representing standard workflows for model discovery and retrieval. - [DATA_EXPOSURE_&_EXFILTRATION]: No sensitive data access or unauthorized network operations were found. References to authentication (
hf auth login) are standard for accessing gated models on Hugging Face.
Audit Metadata