Windsurf MCP Integration: Missing Security Controls Put Users at Risk · Embrace The Red

Windsurf's MCP integration lacks essential security controls, allowing AI agents to automatically invoke tools without user approval. This creates serious vulnerabilities to prompt injection attacks where malicious instructions in code comments or files can hijack the agent to perform unauthorized actions like accessing private Slack channels. The analysis demonstrates how attackers can exploit this zero-click automation and recommends implementing user-configurable approval workflows for different tool categories based on risk levels.