awt-e2e-testing

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructions provide a command to install external code from a GitHub repository using npx skills add ksgisang/awt-skill. This is a standard installation procedure for this platform.
  • [PROMPT_INJECTION]: The skill processes untrusted external data in the form of YAML test scenarios, which creates a potential surface for indirect prompt injection.
  • Ingestion points: YAML-based declarative test scenarios provided to the Playwright executor.
  • Boundary markers: No specific delimiters or safety warnings for embedded instructions are identified in the provided documentation.
  • Capability inventory: The skill uses Playwright for browser automation, OpenCV and OCR for visual matching, and maintains a local SQLite database for failure patterns.
  • Sanitization: There is no documentation regarding the validation or sanitization of the YAML scenario content prior to execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 11:31 AM
Security Audit — agent-trust-hub — awt-e2e-testing