webapp-testing

Warn

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script 'scripts/with_server.py' utilizes 'subprocess.Popen(shell=True)' to execute commands provided via the '--server' argument, which can be used to run arbitrary shell instructions.\n- [REMOTE_CODE_EXECUTION]: The skill encourages a workflow where the agent generates and executes local Python scripts (Playwright automations) at runtime, which is a form of dynamic code execution.\n- [PROMPT_INJECTION]: The documentation in 'SKILL.md' explicitly instructs the agent 'DO NOT read the source until you try running the script first', which discourages source code inspection and safety verification before execution.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data from web applications:\n
  • Ingestion points: 'examples/element_discovery.py' (lines 15-35) reads DOM content and attributes; 'examples/console_logging.py' (lines 14-17) captures browser console messages.\n
  • Boundary markers: No delimiters or safety instructions are used to separate untrusted web data from the agent's command context.\n
  • Capability inventory: 'scripts/with_server.py' (lines 74-83) executes shell commands via 'subprocess'; Playwright scripts have filesystem access for screenshots and logs.\n
  • Sanitization: There is no evidence of sanitization or validation for the data retrieved from the web applications before it influences agent logic.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 6, 2026, 05:29 PM