generative-art-deployment
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill contains a Python script for batch rendering that utilizes
subprocess.runto execute a local Node.js process. This is a standard workflow for algorithmic art pipelines. - [COMMAND_EXECUTION]: A shell script is provided for Linux kiosk mode, which invokes system utilities such as
xset,unclutter, andchromium-browserto configure display settings for installations. - [INDIRECT_PROMPT_INJECTION]: The rendering pipeline snippets define function signatures that accept file paths and command-line arguments, creating a potential surface for injection if populated with untrusted data.
- Ingestion points: Function parameters in the
batch_renderPython snippet inSKILL.md. - Boundary markers: None present in the code examples.
- Capability inventory: Process execution via
subprocess.runinSKILL.md. - Sanitization: The script uses list-based arguments for
subprocess.run, which provides inherent protection against shell injection attacks.
Audit Metadata