motion-craft
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as SRT files and user briefs, which could theoretically contain malicious instructions.
- Ingestion points: SRT files and user-provided briefs are identified as primary inputs in
SKILL.md. - Boundary markers: The
scripts/read-stage.pyscript uses explicitBEGIN_SOURCEandEND_SOURCEmarkers along with packet hashes to delimit content and ensure integrity during pagination. - Capability inventory: The skill includes scripts for file system discovery (
scripts/find-motion-resources.py) and paginated reading of local documentation, as well as complex mathematical calculations for 3D scenes. - Sanitization: Content is delimited by structural markers rather than sanitized, relying on the agent's ability to distinguish between documentation and instructions.
Audit Metadata