swiss-creative-mode-template
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a
DESIGN.mdfile to configure the layout and styles of the generated HTML template. This external ingestion point creates a surface for potential indirect prompt injection if the source file contains instructions designed to override agent behavior during the generation process. - Ingestion points: The skill reads an active
DESIGN.mdfile to map palette, typography, and layout decisions into CSS variables as described in theSKILL.mdworkflow. - Boundary markers: The workflow does not specify the use of delimiters or 'ignore embedded instructions' warnings when processing the design file content.
- Capability inventory: The skill is granted
file_writecapabilities to emit theindex.htmlartifact. - Sanitization: There is no mention of validation or sanitization routines for the data extracted from the design file before it is interpolated into the generated code.
Audit Metadata