AGENT LAB: SKILLS

gemini-computer-use

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Prompt Injection (LOW): Indirect Prompt Injection surface detected. The agent interprets content from external websites which could contain malicious instructions.
  • Ingestion points: Screenshots and URL content of arbitrary web pages via Playwright as described in SKILL.md.
  • Boundary markers: No explicit prompt delimiters or 'ignore embedded instructions' warnings are provided in the references.
  • Capability inventory: High-impact actions including click_at, type_text_at, and key_combination are supported as listed in google-computer-use.md.
  • Sanitization: No sanitization or filtering of ingested web data is specified in the skill files.
  • External Downloads (SAFE): The skill requires the installation of standard libraries and browser binaries.
  • Evidence: pip install google-genai playwright and playwright install chromium.
  • [TRUST-SCOPE-RULE]: google-genai is a package from a trusted organization (Google).
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:13 PM