antislop-decks
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted client data and project files to generate presentations across various formats (pptx, Google Slides, HTML).
- Ingestion points: Reads client content and optional DESIGN.md files from the project folder.
- Boundary markers: None explicitly defined in the provided instructions to isolate user content from system instructions.
- Capability inventory: Capable of writing files (.md, .pptx, .html) and interacting with the Google Slides API.
- Sanitization: No specific sanitization or validation of the input content is described.
- [SAFE]: The skill instructions focus entirely on stylistic and structural discipline for presentation design. It explicitly avoids using external MCP tools or network services, relying instead on local library integration for deck generation.
Audit Metadata