skills/openai/plugins/ngs-runtime-env/Gen Agent Trust Hub

ngs-runtime-env

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Command Execution]: The skill executes various system commands and package managers (e.g., conda, pip, docker, nextflow) to verify and install bioinformatics tools. This is a standard part of sequencing pipeline management.
  • [External Resource Retrieval]: The skill facilitates downloading software and reference databases from public repositories such as Bioconda and PyPI. These are well-known resources in the scientific community.
  • [Script Generation]: The included Python scripts generate shell files (install_commands.sh and resource_setup_commands.sh) for users to review. This practice provides a manual verification step before system changes occur, ensuring that automated actions remain under user control.
  • [Data Ingestion Surface]: The skill processes inputs such as file paths and pipeline names to generate setup commands. While these inputs influence generated code, the skill mitigates risk by requiring explicit user confirmation (e.g., via the --yes flag or environment variables) and emphasizing the emission of reviewable plans before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 05:33 AM
Security Audit — agent-trust-hub — ngs-runtime-env