create-sticker
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by interpolating raw user input into the prompt sent to the LLM.
- Ingestion points: User input enters the script through the
descriptionargument inscripts/create_sticker.py. - Boundary markers: Absent. The user-provided text is directly appended to the system prompt string (
prompt = f"{CHARACTER_BASE}\nAction/Pose: {desc}") without delimiters or instructions for the model to ignore embedded commands. - Capability inventory: The skill possesses network access via the
google-genaiSDK and file system write access to the user's home directory (~/stickers/) to save generated images. - Sanitization: No validation or sanitization is performed on the description input before it is used for prompt generation.
Audit Metadata