parabricks
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a diagnostic script
scripts/check_parabricks_runtime.pythat executes local system commands such asnvidia-smianddockerto assess the host's runtime readiness. These commands are implemented using safe subprocess execution patterns (argument lists) rather than shell strings, which effectively mitigates the risk of command injection. - [EXTERNAL_DOWNLOADS]: The diagnostic script can be configured to pull Docker images from the official NVIDIA Container Registry (
nvcr.io) or a user-provided image to verify that the environment can successfully launch Parabricks containers. This is standard functionality for the skill's purpose and targets trusted infrastructure. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data such as file paths, reference builds, and software version tags to generate
pbruncommand templates. While this represents a surface for indirect prompt injection (e.g., via malicious file names), the skill includes explicit guardrails instructing the agent to generate conservative Docker commands with explicit mounts and validate all paths and outputs before finalized generation.
Audit Metadata