chrome-extension

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: LOW
Full Analysis
  • Category 1: Prompt Injection (SAFE): No instructions were found that attempt to override system prompts, bypass safety filters, or disclose internal instructions.
  • Category 2: Data Exposure & Exfiltration (SAFE): The skill focuses on browser extension development and does not reference sensitive local files (~/.ssh, .env) or unauthorized data exfiltration patterns. It correctly identifies broad host permissions as sensitive.
  • Category 3: Obfuscation (SAFE): No encoded strings (Base64), zero-width characters, or homoglyphs were detected.
  • Category 4: Unverifiable Dependencies & RCE (SAFE): The skill does not perform external package installations or execute remote code. External links point to official Google developer documentation.
  • Category 5: Privilege Escalation (SAFE): No commands related to system-level privilege escalation (sudo, chmod) are present.
  • Category 6: Persistence Mechanisms (SAFE): No host-level persistence mechanisms are established. Browser service workers are mentioned only in the context of standard extension architecture.
  • Category 8: Indirect Prompt Injection (SAFE): The skill is instructional and does not define ingestion points for untrusted external data that would influence agent reasoning.
  • Category 10: Dynamic Execution (SAFE): While it discusses chrome.scripting.executeScript, this is provided as technical guidance for browser extension functionality, not for executing arbitrary code on the user's host machine.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 17, 2026, 12:35 AM