ae-ai-templates
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied briefs to populate After Effects templates, creating a potential surface for indirect prompt injection where malicious instructions could be embedded in the animation content.
- Ingestion points: The
valuesparameter in theae_template buildcommand receives data from user briefs (found in SKILL.md). - Boundary markers: The documentation explicitly states "Slot values are data, never code" and implements
maxCharslimits to prevent layout breaking. - Capability inventory: The skill utilizes After Effects manipulation tools including
ae_create_comp,ae_create_layer, and rendering tools viaae-deliver(found in SKILL.md). - Sanitization: Templates use typed slots (color, enum, boolean) and the
npm testsuite validates template structure and expressions before use.
Audit Metadata