cli-anything-obs-studio

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation includes instructions to install the cli-anything-obs_studio package via pip. This is a standard dependency installation from the official Python package registry for the vendor's own tool.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of the cli-anything-obs_studio command-line utility. These commands are used for local project management, scene editing, and configuration, all of which are consistent with the tool's primary purpose.
  • [INDIRECT_PROMPT_INJECTION]: The tool processes JSON-based project files, which represent an ingestion surface for external data. However, the data is highly structured and intended for local OBS configuration, presenting a negligible risk in this context.
  • Ingestion points: User-provided project files (e.g., project.json) and command-line parameters for scene and source properties.
  • Boundary markers: Not explicitly defined within the markdown documentation, as the tool relies on structured JSON parsing.
  • Capability inventory: The skill enables local file writing (creating and saving project files) and modification of OBS Studio configurations via the CLI wrapper.
  • Sanitization: Input validation is expected to be managed by the underlying CLI tool when parsing the JSON schema.
  • [SAFE]: No indicators of credential exfiltration, persistence, privilege escalation, or obfuscation were found. The skill operates within its defined scope of a configuration management tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 05:21 PM
Security Audit — agent-trust-hub — cli-anything-obs-studio