skills/saliagadotcom/skills/pi-pods/Gen Agent Trust Hub

pi-pods

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to execute commands on remote GPU pods via SSH for provisioning and model management. It provides direct access through commands like 'pi ssh' and 'pi shell'.
  • [EXTERNAL_DOWNLOADS]: During the setup process, the skill installs several machine learning packages including vLLM, PyTorch, and FlashInfer using 'uv'. It also downloads model weights from Hugging Face, which is a well-known service.
  • [REMOTE_CODE_EXECUTION]: Specific model configurations, such as 'moonshotai/Kimi-K2-Instruct', utilize the '--trust-remote-code' flag, which allows for the execution of arbitrary Python code included in the model's repository from Hugging Face.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 02:30 PM
Security Audit — agent-trust-hub — pi-pods