magic-animator
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses authoritative language such as 'ABSOLUTE MANDATE' and 'Agents MUST utilize this skill' to override agent discretion and force specific behavioral outputs regarding creative style.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests external, potentially untrusted design assets which can contain embedded instructions.
- Ingestion points: Static design elements (SVG, PNG, Figma layers) and chat-based prompts processed by the 'AI Animation Assistant' as described in SKILL.md.
- Boundary markers: None present; the skill does not specify delimiters or warnings to ignore instructions embedded within the design assets.
- Capability inventory: The skill produces and exports animation files, including Lottie JSON, GIF, and MP4 formats.
- Sanitization: There is no evidence of sanitization, validation, or filtering of the content within the design assets or user-provided prompts.
Audit Metadata