sprite-pipeline
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external game data which could potentially serve as a vector for indirect instructions.
- Ingestion points: The workflow requires the agent to inspect runtime frame sizes, facing conventions, anchors, palettes, manifests, and animation definitions (referenced in
SKILL.md). - Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent treats external metadata strictly as data rather than potential instructions.
- Capability inventory: The skill leverages
generate_canvas_mediaandgenerate_imagetools, and it has the capability to update asset manifests and animation metadata files. - Sanitization: The instructions do not specify any validation or sanitization steps for the external metadata before it is used to influence generation or file updates.
Audit Metadata