modelopt-accessing-mlflow
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The troubleshooting section contains a command to install the 'uv' tool:
curl -LsSf https://astral.sh/uv/install.sh | sh. This pattern downloads and executes a shell script directly from Astral's official domain. - [EXTERNAL_DOWNLOADS]: The skill references an external Model Context Protocol (MCP) server, 'mlflow-mcp', hosted on GitHub (github.com/kkruglik/mlflow-mcp), which is required for the skill's functionality.
- [COMMAND_EXECUTION]: The skill provides examples of executing Python code via
uv runto perform data analysis and metric comparisons on fetched MLflow data. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external MLflow artifacts, such as
results.yml,config.yml, andeval_factory_metrics.json. These files represent an attack surface where untrusted data from evaluation runs could potentially contain instructions aimed at the agent.
Audit Metadata