RingCentral, Inc. (NYSE: RNG) announced a new plugin for ChatGPT and model context protocol (MCP) connectors for Claude, making voice, SMS, and chat data from its platform accessible within large language models (LLMs).

The MCP connectors are available with all paid RingEX plans and support voice, SMS, and chat data. The ChatGPT plugin is available through RingCentral's App Gallery and the ChatGPT Directory.

According to the company, the integrations allow business users to combine RingCentral conversation data with LLM reasoning to automate tasks such as prioritizing sales pipelines, sending SMS follow-ups based on abandoned shopping carts, flagging urgent support issues, and generating daily task lists from calls and messages.

RingCentral said the connectors include role-based and individual access controls, maintaining identity-aware permissions consistent with a user's existing settings, including the ability to limit LLM access to specific data types.

Cam Lothridge, a franchise owner of OrangeTheory Fitness in Warner Robins, Georgia, said the ChatGPT plugin "increased our lead response rate by 50 percent" and reduced the time staff spent reviewing customer conversations from 45 minutes to under five minutes per shift.

Kira Makagon, President and COO of RingCentral, said in a statement that MCP "makes that interaction context available directly in the LLMs that businesses already use, so they can securely utilize their own data to drive outcomes."

Michelle Morgan, Research Director at IDC, said open protocols like MCP allow conversation context to move to "wherever the work happens, instead of requiring a new integration for every AI platform."

RingCentral described the MCP connectors as the first step in a broader agent platform roadmap.