SwipeFlow MCP Server enables human-in-the-loop approvals for AI agents over the Model Context Protocol, allowing agents to request asynchronous human review of work (content, code, messages, spending) and resume once a decision is made, with stateless, durable task handles and synchronous fallbacks for clients without Tasks extension support.
Enables AI agents to pause execution at critical decision points and request human review before proceeding. Provides tools for creating interrupts, polling for decisions, and managing approvals through a simple REST API interface.