playwright-mcp

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads and installs the @playwright/mcp package and required browser binaries (e.g., Chromium) from the official npm registry and Playwright distribution servers during setup.
  • [COMMAND_EXECUTION]: Utilizes npm and npx commands to install the MCP server, manage browser installations, and execute the automation server in the local environment.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it reads and processes data from external, untrusted websites.
  • Ingestion points: External data enters the agent context through browser_get_text, browser_snapshot, and browser_evaluate operations (SKILL.md).
  • Boundary markers: The skill instructions do not specify any boundary markers or instructions to the agent to ignore embedded commands within the extracted web content.
  • Capability inventory: The skill allows the agent to navigate to any URL, interact with forms, execute arbitrary JavaScript in the browser, and extract page data (SKILL.md, examples.py).
  • Sanitization: There is no evidence of sanitization or filtering of the web content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 05:35 PM
Security Audit — agent-trust-hub — playwright-mcp