spin-design
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill functions as a prompt generator that interpolates untrusted data from the conversation history and project files (e.g., manuscripts, copy) into a new DESIGN-PROMPT.md. This creates a surface where instructions in the source material could influence the subsequent design session.
- Ingestion points: Conversation context and local project files (manuscripts, copy, asset descriptions) as specified in SKILL.md.
- Boundary markers: SKILL.md does not provide instructions for using delimiters or boundary markers in the generated prompt files.
- Capability inventory: File system access for reading existing assets and writing new files within the project subfolder.
- Sanitization: No explicit sanitization or filtering of interpolated content is required by the instructions.
- [COMMAND_EXECUTION]: The handoff guidelines in SKILL.md describe a workflow utilizing headless Chrome with the --print-to-pdf flag to render previews of design artifacts. This is a standard procedure for converting HTML designs into static documents.
Audit Metadata