ppt-generator
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a bash command to assemble PPTX files using
python ~/.claude/skills/ppt-generator/scripts/assemble_pptx.py. The command incorporates a project name placeholder that is derived from user input. This pattern creates a risk of command injection if the project name contains shell metacharacters and is not properly sanitized by the execution environment. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by interpolating untrusted user data into Python script templates.
- Ingestion points: User-provided PPT titles, slide content, and project names are processed in the
SKILL.mdworkflow. - Boundary markers: No delimiters or protective warnings are used when placing user input into Python variable assignments.
- Capability inventory: The skill performs filesystem writes to
~/ppt-output/and executes shell commands via theassemble_pptx.pyscript. - Sanitization: The instructions do not specify any validation or escaping mechanisms for the user-supplied content before it is used in code or shell execution.
- [EXTERNAL_DOWNLOADS]: The skill's configuration references well-known services such as Google, ModelScope, and Volcengine for image generation tasks. These are established technology providers and are treated as safe sources for the intended functionality.
Audit Metadata