bitwarden

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow reads public web pages from bitwarden.com via an MCP server, which constitutes fetching vendor-published or searchable web content rather than monitoring an incoming outsider-submission channel.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The skill configures and relies on a remote MCP server hosted at unbrowse.ai, which provides the compiled operational tools and behavior-selecting configuration for the agent, representing a third-party unofficial operational dependency.