pitch-deck-agent
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input describing business details, traction, and goals which is subsequently used to construct image generation prompts and slide content.
- Ingestion points: Interactive user responses to Question Flow (Q1-Q8) in Step 1.
- Boundary markers: There are no explicit delimiters or instructions to the agent to ignore embedded commands within the user-provided data.
- Capability inventory: The skill is designed to execute local Python scripts (
gemini.py) via subprocess and write output files to the local filesystem. - Sanitization: No sanitization, escaping, or validation of user-provided business descriptions is performed before they are interpolated into the generated output or command arguments.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute a local Python script (
python3 ${SKILL_PATH}/skills/image-generation/scripts/gemini.py) to generate visuals, incorporating generated prompts based on user input into the command-line arguments.
Audit Metadata