webapp-testing

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution Utility: The scripts/with_server.py utility manages local development servers and test scripts by executing shell commands. It uses subprocess.Popen with shell=True to support complex initialization (e.g., cd backend && npm start) and subprocess.run for the final execution step. This is a core feature intended for server lifecycle management and test automation in developer environments.
  • Browser Automation Surface: The skill interacts with web applications via Playwright, creating a surface for potential indirect prompt injection. 1. Ingestion points: DOM content and console logs are read into the agent context in examples/element_discovery.py and examples/console_logging.py. 2. Boundary markers: The skill does not currently employ specific delimiters to isolate processed web content from instructions. 3. Capability inventory: The skill possesses the ability to execute shell commands via with_server.py and write files to designated output paths like /mnt/user-data/outputs/. 4. Sanitization: Content retrieved from web pages is processed without explicit filtering or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 02:53 PM
Security Audit — agent-trust-hub — webapp-testing