frontend-slides
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to open the generated presentation files in the user's browser, using
open(macOS),xdg-open(Linux), orstart(Windows). - [EXTERNAL_DOWNLOADS]: To handle PowerPoint conversions, the skill utilizes the
python-pptxlibrary and suggests installing it via Python's package manager if it is not already present in the environment. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: Extracted text, images, and notes from user-provided
.pptand.pptxfiles. - Boundary markers: There are no specific delimiters or instructions provided to the agent to treat the converted content as untrusted data or to ignore embedded instructions.
- Capability inventory: The skill can execute shell commands (
open,python3) and write files to the local system. - Sanitization: The instructions do not define sanitization or filtering procedures for the extracted PowerPoint content before it is included in the final HTML presentation output.
Audit Metadata