seedance-20
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: Comprehensive analysis of the skill's instructions, metadata, and supporting scripts reveals no attempts at prompt injection, data exfiltration, or obfuscation. The skill adheres to best practices for state management and user privacy.
- [COMMAND_EXECUTION]: The repository contains several Python scripts (e.g.,
scripts/validate_skills.py,scripts/install_codex_skill.py) and a test suite. These scripts use thesubprocessandshutilmodules to perform local file system operations and run validation checks. These actions are clearly documented and are necessary for the development and installation of the filmmaker OS. - [EXTERNAL_DOWNLOADS]: The evaluation script
scripts/eval_run.pymakes outbound network calls toapi.anthropic.com. This is used to programmatically score generation prompts against an internal rubric. This functionality targets a well-known service and is an intended part of the quality assurance workflow.
Audit Metadata