ppt-image-first-workflow
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires cloning source code from a third-party GitHub repository (github.com/NyxTides/ppt-image-first.git) during the installation process.
- [COMMAND_EXECUTION]: The Python code uses subprocess.run to invoke system commands (open or xdg-open) to launch the default web browser for viewing local slide previews.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user materials to generate presentation content, creating a vulnerability to indirect prompt injection.
- Ingestion points: User materials such as reports and notes are collected via the materials field in the intake form.
- Boundary markers: The skill logic does not implement specific delimiters or instructions to prevent the model from following commands embedded within user-provided text.
- Capability inventory: The skill possesses network access for image generation, filesystem write permissions for output, and shell command execution for launching the browser.
- Sanitization: No sanitization or filtering mechanisms are applied to external user data before it is incorporated into image generation prompts.
Audit Metadata