scenescape-setup
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes subprocess and docker commands to orchestrate the deployment of SceneScape. These calls are used to manage containers, fetch assets via git, and execute helper scripts. The usage is consistent with the skill's primary purpose and employs safe argument passing techniques.
- [EXTERNAL_DOWNLOADS]: It fetches additional code components and detection models from official project repositories on GitHub and well-known images from Docker Hub. These resources are required for functionality and originate from the project's own infrastructure or established registries.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied RTSP URLs, camera IDs, and video files. It includes validation logic in scripts like deploy_inputs.py to ensure inputs meet expected formats and constraints (e.g., URL schemes, character restrictions) before they are used in the deployment process, effectively mitigating potential injection risks.
- [SAFE]: Secret management follows security best practices. The skill generates random passwords for the database and superuser at deployment time and stores them in a .env file with restricted (0600) permissions to prevent unauthorized access by other users on the host system.
Audit Metadata