Miro MCP Integration

Connect OpenClaw agents to Miro via Model Context Protocol (MCP). Use when generating diagrams, visualizing code, brainstorming board layouts, or integrating...

This skill appears to do what it says: connect an OpenClaw agent to Miro via MCP and read/write board content after you authenticate. Before installing/using: (1) Confirm which Miro team you select during OAuth — the MCP app is team-scoped and selecting the wrong team blocks access; (2) Review OAuth scopes requested by your client and avoid granting write permissions if you only need read access; (3) Be aware that workflows that ask the agent to 'Analyze ~/dev/myapp' or to read GitHub repos require the agent (or client) to access local files or repository content — only allow that when you trust the agent and data involved; (4) context_get calls consume Miro AI credits and are rate-limited, so expect costs/limits for heavy reads; (5) Optional developer tooling suggestions (e.g., npm install -g @modelcontextprotocol/inspector) are not required and should be installed only if you trust the source. If you want a stricter privacy posture, require read-only scopes and avoid prompts that ask the agent to read local or sensitive repositories.

Type: OpenClaw Skill Name: miro-mcp Version: 1.0.0 The bundle provides comprehensive documentation and instructions for integrating OpenClaw agents with Miro via the Model Context Protocol (MCP). It utilizes official Miro endpoints (mcp.miro.com and api.miro.com) and implements standard OAuth 2.1 authentication with team-scoped access controls. While the skill allows agents to read local codebases and board content, these actions are strictly aligned with the stated purpose of generating architecture diagrams and implementation guidance. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found across the SKILL.md or reference files.