gpt-image-slide
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content, including visual references and source files, to generate slide plans and prompts.
- Ingestion points: Data enters the workflow from reference slide images, source files, and user deck requests as described in SKILL.md.
- Boundary markers: The instructions lack specific delimiters or warnings to the agent to ignore potentially malicious instructions embedded within the source files.
- Capability inventory: The workflow involves writing multiple files (DESIGN.md, slide_plan.json, slide_prompts.json) and generating image files in the local workspace.
- Sanitization: There is no evidence of sanitization, validation, or filtering of external content before it is interpolated into the slide generation process.
Audit Metadata