KASHIKA Inc. (Headquarters: Ota Ward, Tokyo; Representative Director: Kenta Ozawa) officially released MCP (Model Context Protocol) on July 1, 2026. This feature allows users to operate sales functions using natural language with AI assistants like Claude within the AI sales tool "AI Apo-rou-kun". In conjunction with this release, MCP has been standardized in the basic Standard Plan, making it available to contracted companies at no additional cost and free for one account. This enables users to proceed with "company search → list creation → message drafting → form/email sending → response visualization," which previously required operating a management screen, simply by speaking to their AI assistant. The role of sales representatives will shift from "manual labor" to "supervising the AI's activities and making judgments." [Background of Development] - "We want to do form sales, but we don't have the staff to manage it." More companies are engaging in "form sales" and "email sales" as means of new customer acquisition. However, feedback from the field consistently includes the following concerns: - Lack of personnel to continuously handle company searches, list creation, message drafting, sending, and response confirmation. - Even after introducing tools, template updates and operations become a burden, hindering the sales PDCA cycle. - Despite the willingness to engage, daily workload prevents execution. In essence, the fundamental issue is not "not knowing how to do it," but rather "who will continue to execute it." Now that AI agents have reached a practical stage, the starting point for this release was the question of whether this "execution" itself could be entrusted to AI. What MCP Can Do - "Sales Execution" Opens Up to AI AI Apo-rou-kun's MCP provides the functions that constitute sales execution as "components" that AI can directly call. Function Capabilities Company and Signal Search Extract prospective companies based on industry, size, and area, as well