tensorrt-llm
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate deployment and optimization guidelines for the NVIDIA TensorRT-LLM library. All referenced external resources, including Docker images and GitHub documentation, originate from well-known and trusted organizations.
- [EXTERNAL_DOWNLOADS]: The instructions include pulling official Docker images (nvidia/tensorrt_llm) and installing packages from the standard Python Package Index (PyPI). These operations are standard for the described use case and do not involve untrusted or malicious sources.
- [COMMAND_EXECUTION]: The shell commands provided in the documentation are restricted to standard model serving (trtllm-serve) and local API testing via curl. There is no evidence of unauthorized or hidden command execution.
- [CREDENTIALS_UNSAFE]: The skill demonstrates safe practices for handling sensitive information by using environment variables (e.g., $API_KEY) in serving examples rather than providing hardcoded secrets.
Audit Metadata