ltx-2-5
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for transforming untrusted "shot briefs" into structured prompts, which creates a surface for indirect prompt injection.
- Ingestion points: Untrusted input from external "shot briefs" is used as the primary data source for prompt generation.
- Boundary markers: Absent; the instructions do not include the use of delimiters or directives to ignore instructions embedded within the source briefs.
- Capability inventory: The skill directs the agent to write generated content to the file system at
04_generation/prompts/ltx-2-5/<shot-id>.md. - Sanitization: Absent; the skill does not specify any sanitization or validation of the input data and explicitly instructs the agent to preserve certain elements from the source brief "without paraphrasing," facilitating the potential propagation of malicious instructions.
Audit Metadata