playwright-cli

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Dynamic Code Execution Considerations: The skill utilizes run-code and eval commands to execute arbitrary JavaScript within the browser environment. This is a core feature for advanced automation, though it requires the agent to be mindful of the code it generates and executes.
  • Indirect Prompt Injection Surface: The skill is designed to interact with and extract data from external, untrusted web pages. The documentation explicitly warns that tool definitions provided by pages (via WebMCP) should be treated as untrusted input to mitigate the risk of indirect injection.
  • Sensitive Browser Data Access: The toolset includes commands to read and modify cookies, localStorage, and session state. The instructions provide guidance on safe handling, such as advising against committing storage state files containing authentication tokens to version control.
  • External Software Installation: The documentation guides the user to install official Playwright packages from the NPM registry. These are well-known resources provided by the tool's vendor and are part of the standard setup process.
  • Visual Data Capture: Capabilities for taking screenshots and recording videos are included. While useful for debugging and reporting, these features can capture any information visible on the automated pages.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 03:46 PM
Security Audit — agent-trust-hub — playwright-cli