google-ai-studio-tts
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from users (scripts, dialogue, and narration text) and interpolates it into the Google AI Studio web application using browser automation tools.\n
- Ingestion points: User-provided script text provided in the prompt is mapped to form fields in
SKILL.mdandreferences/selectors.md(e.g.,textarea[aria-label=\"Enter a prompt\"]).\n - Boundary markers: The instructions do not define explicit boundary markers or sanitization steps for the user-supplied content before it is entered into the web form.\n
- Capability inventory: The skill utilizes browser control tools (
claude-in-chrome) capable of navigating, inputting form data, and executing arbitrary JavaScript within the context of the user's browser session onaistudio.google.com.\n - Sanitization: There is no evidence of explicit input validation or escaping for the text content entered into the browser DOM beyond standard tool-level handling.
Audit Metadata