ltx2
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were detected. The skill implementation follows standard patterns for CLI-based AI tools.
- [CREDENTIALS_UNSAFE]: The skill directs users to utilize secure secret management via .env files and official CLI commands for API tokens, which is a recommended practice.
- [COMMAND_EXECUTION]: The skill uses the uv package runner to execute local scripts. These commands are localized to the provided toolkit and do not involve untrusted remote sources.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text and image inputs to generate video content. 1. Ingestion points: --prompt and --input parameters in tools/ltx2.py. 2. Boundary markers: None defined in the documentation. 3. Capability inventory: Local script execution via uv run. 4. Sanitization: Not specified, but the impact is confined to the generated media output.
Audit Metadata