skills/av/skills/pull-llamacpp-model/Gen Agent Trust Hub

pull-llamacpp-model

Pass

Audited by Gen Agent Trust Hub on Apr 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the harbor command-line interface to modify system configurations via harbor config set. These commands are used to change and subsequently restore the container image used for model inference, which is the stated purpose of the skill.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch a Docker image from ghcr.io/ggml-org/llama.cpp:server. This repository belongs to the official organization responsible for the llama.cpp project, a well-known service in the AI community.
  • [EXTERNAL_DOWNLOADS]: The skill references and restores a custom Docker image kyuz0/amd-strix-halo-toolboxes:rocm-7.2. The instructions explicitly state this image is required for specific AMD hardware support (Strix Halo / gfx1151) that is not available in official ROCm builds.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 4, 2026, 02:26 PM
Security Audit — agent-trust-hub — pull-llamacpp-model