frontend-slides
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use platform-specific shell commands such as 'open', 'xdg-open', and 'start' to view the generated presentation files.
- [EXTERNAL_DOWNLOADS]: It suggests the installation of the 'python-pptx' library from a public package registry to support its PowerPoint conversion feature.
- [PROMPT_INJECTION]: The skill ingests untrusted data from external .ppt and .pptx files, creating a potential surface for indirect prompt injection if those documents contain malicious instructions.
- Ingestion points: Data extraction from user-provided PowerPoint files via the conversion workflow in 'SKILL.md'.
- Boundary markers: Absent; there are no instructions to use specific delimiters or to ignore instructions found within the ingested slide content.
- Capability inventory: Shell command execution (python3, file openers), file system write access for HTML and assets, and potential browser-based validation.
- Sanitization: Absent; the skill does not explicitly provide instructions for sanitizing or validating text extracted from external documents before it is used in code generation.
Audit Metadata