Vanilla MCP connects a compatible AI assistant to your Vanilla community. Your team can ask questions about community activity, find content that needs attention, and use available tools to take action without manually collecting links and reports first.
MCP in brief
MCP stands for Model Context Protocol. It is a standard way for an AI assistant to connect to another service's tools. Think of Vanilla MCP as the bridge between your chosen AI assistant and your Vanilla community: the assistant asks for an approved tool, Vanilla returns the information or carries out the action the signed-in user is allowed to perform, and the assistant uses that result to answer you. MCP is a connector, not an AI model or a separate copy of your community database.
Vanilla MCP is currently in beta. Capabilities may change as we learn from participants. The tools you can use also depend on your community configuration and your Vanilla role and permissions. Check the results and review proposed changes before acting on your community.
Want to be part of the beta? Visit the Vanilla MCP Beta Group to learn more and join the conversation.
Sign in to connect
Each person who uses Vanilla MCP must complete its authentication page when connecting their AI tool. In the current beta, this is a separate Vanilla email/password sign-in rather than your community's SAML SSO page. If it is your first time, you may need to create that sign-in and verify an emailed code. SSO support is planned for general availability; it is not part of the current beta. Do not use a shared account: each person's connection uses their own Vanilla role and permissions.
Explore community content
- Find relevant discussions, questions, and knowledge base content.
- Summarize an active thread or investigate what members are asking about a topic.
- Find unanswered questions and conversations that may need a response.
- Explore ideas and feedback to see what customers are raising.
For example, you could ask: “Which questions from the past two weeks have no replies?”
Focus your daily review
- Review reported posts and content awaiting moderation.
- Look for possible spam or conversations that may need closer attention.
- Find active or recently revived discussions.
- Identify open escalations that need follow-up, where that workflow is enabled.
For example, you could ask: “What needs my attention in the community today?”
Find opportunities to engage
- Identify new or active contributors worth welcoming or recognizing.
- Review community content that may be useful to feature or organize.
- Explore available activity and performance signals to guide your next question.
For example, you could ask: “Which new members joined recently and have not posted yet?”
Take action with the right permissions
Depending on your role and the tools enabled for your community, Vanilla MCP can help you follow up on what you find. Examples include replying to content, organizing content with tags or collections, handling moderation work, and recognizing members with badges. Review the proposed action in your AI assistant before you run it. Your existing Vanilla permissions continue to govern what you can see and do.
Security and data handling
Vanilla MCP uses OAuth 2.0 for individual authentication and TLS to protect data in transit. It calls Vanilla's existing API on behalf of the signed-in user; Vanilla applies that person's existing roles and permissions. The MCP server retrieves community content as needed and does not retain that content after responding to a request. The AI assistant runs in the provider your organization chooses, so review that provider's settings and policies for how it handles content you send there. MCP actions are logged for audit. For more about authentication, access controls, and data handling, see Vanilla MCP security and data handling (Beta).
Get started
Start with Model Context Protocol (MCP) in Vanilla for an overview of how the connection and permissions work. If your team uses Microsoft Copilot Studio, see Connect Vanilla Community to Microsoft Copilot Studio via MCP.
Beta participants can find onboarding, examples, updates, and feedback discussions in the Vanilla MCP Beta Group.