frontend-slides
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from user-provided PowerPoint files (.pptx) and notes which are then interpolated into the final HTML presentation. An attacker could potentially embed malicious instructions or scripts within the PowerPoint content.
- Ingestion points: Content is extracted from
.pptxfiles via theextract_pptxfunction inSKILL.md. - Boundary markers: The skill does not specify the use of delimiters or 'ignore' instructions to separate extracted content from the agent's logic.
- Capability inventory: The skill can write files, execute Python scripts, and run shell commands (e.g.,
open [filename].html). - Sanitization: There is no evidence of sanitization or escaping of the extracted PowerPoint content before it is rendered in the generated HTML.
- [DYNAMIC_EXECUTION]: The skill generates and executes local Python code to extract contents from PowerPoint files and creates HTML/JavaScript files that are intended to be opened in the user's browser.
- [EXTERNAL_DOWNLOADS]: The skill fetches typography resources from Google Fonts and Fontshare to style the presentations. These are well-known and established services for web development.
Audit Metadata