gpt-image-slide-plan
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 input from external sources, creating a potential surface for indirect prompt injection.
- Ingestion points: The instructions specify using
DESIGN.mdand "user's files" as the primary content pool for planning (SKILL.md). - Boundary markers: There are no explicit delimiters or boundary instructions provided to help the agent distinguish between its core instructions and the data being processed.
- Capability inventory: No high-risk capabilities like network access, command execution, or file writing are present in the skill; it only generates a JSON output.
- Sanitization: The skill lacks instructions for sanitizing or validating content retrieved from external files before it is incorporated into the slide planning process.
Audit Metadata