deepstream-generate-config

Pass

Audited by Gen Agent Trust Hub on Oct 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed to generate DeepStream configuration files through a set of bundled Python scripts (render.py, validate.py, etc.). My analysis confirms that these scripts do not perform any malicious actions such as network exfiltration, credential harvesting, or unauthorized file access.
  • [SAFE]: Data ingestion is handled securely. The Python scripts use yaml.safe_load() to process user-supplied specifications, which prevents arbitrary code execution during the parsing of YAML data.
  • [SAFE]: The skill implements robust path validation. Specifically, the _sidecar_output_path function in render.py uses path resolution and relative-path checks to ensure that generated files cannot be written outside the intended output directory, effectively mitigating path traversal risks.
  • [SAFE]: The skill explicitly adheres to the principle of least privilege. The instructions in SKILL.md strictly forbid the use of sudo or other privileged commands and caution against writing to root-owned directories unless explicitly requested by the user for standard sample locations.
  • [SAFE]: No obfuscation, prompt injection attempts, or persistence mechanisms were detected. All external dependencies (e.g., PyYAML) are standard, version-pinned packages from official registries.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 31, 2026, 09:41 PM