Miro has reported more than 16 million calls to its MCP server since its launch in February 2026, with usage more than doubling since May. In September alone, MCP calls are on track to reach 6 million.
Teams increasingly rely on AI to accelerate work—from discovery and planning to design and code. Yet most AI interactions happen in chat windows or static outputs that exist separately from the team’s actual workspace. Miro’s MCP server changes this by giving users a way to make the output from their AI tools visual, immediately editable, and accessible to the whole team on a shared canvas.
Miro user data shows Claude Desktop and Claude Code represent almost two-thirds of usage in August (65%). ChatGPT has become the second-largest client, growing 71% in users between August and September. In total, 15 different AI clients have now connected via the MCP integration; others include Microsoft Copilot, Gemini Enterprise, and Grok Build. Further analysis of the data also demonstrates that the Canvas is being leveraged as both an input and output for AI agents. 57% of August calls were Miro-to-code, where an agent reads a board and acts on it, and 41% were code-to-Miro, where agents generate boards, diagrams and layouts directly.
The data also indicates that the demographic of users adopting MCP has shifted. MCP has jumped the developer niche. Users in engineering and developer roles account for under a quarter of MCP users—product, design, marketing, operations and project management together outnumber engineers roughly two to one.
“The explosive user adoption of Miro’s MCP server is showing the true demand for bringing your agents closer to the work teams are actively doing,” said Jeff Chow, Chief Product and Technology Officer at Miro. “By leveraging MCP tools with Miro, teams can visualize complex ideas across multiple formats on a collaborative canvas. What’s most exciting is the growth we’re seeing across all disciplines, showing Miro is well positioned to be the collaboration layer in the agentic era.”
“The Miro MCP server changed how our product team thinks and ships, said Dominic Le Garsmeur, Chief Product Officer at Fincore Ltd. “It gives our PMs the freedom to be bold on the canvas knowing Claude will help turn that thinking directly into the requirements our developers build from. Better ideas reach engineering faster.”
Teams are adopting Miro as the central interface for AI-assisted work. With 16 million MCP calls since its launch and adoption spreading across all job functions, the pattern is clear: AI agents need a workspace that supports how teams actually work—visual, shared, and endlessly editable.
