ltx2
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructs users to store sensitive information such as the Modal endpoint URL in a
.envfile, which is a standard security best practice for local development to prevent hardcoding secrets in source code. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-supplied text prompts to generate video content. While this is the intended functionality, it represents an attack surface where malicious data could attempt to influence the agent's behavior.
- Ingestion points: The
--promptargument in the execution oftools/ltx2.pyas described inSKILL.md. - Boundary markers: The documentation does not specify the use of delimiters or 'ignore' instructions for the interpolated prompt data.
- Capability inventory: The skill executes shell commands to run a Python script (
tools/ltx2.py) which makes network requests to a remote Modal endpoint. - Sanitization: No explicit sanitization or validation of the prompt string is mentioned in the provided documentation.
Audit Metadata