# Vexa > Send a bot into a Google Meet, Teams or Zoom call and read its live, speaker-labelled transcript from your agent. Vexa's hosted MCP server lets your agent send a Vexa bot into a Google Meet, Microsoft Teams or Zoom call. The bot joins the meeting as a participant everyone can see and transcribes it with speaker labels as it happens, and your agent reads the transcript during the meeting or after it. What you can do with it - Send a bot to a meeting from its link, check on it and stop it when you're done - Read the transcript live or afterwards, with who said what and when - List your meetings and recordings - Search what was said across your meetings, and read the in-call chat - Give a meeting a title and your own tags, such as a CRM or ticket ID - Have the bot say something out loud in the meeting, only when you ask it to How it connects - Address: https://api.cloud.vexa.ai/mcp (streamable HTTP). - Sign in at vexa.ai with a Google account and copy your API key from your account page. Your agent sends it as an Authorization: Bearer header. vexa.ai/connect can also set your agent up for you. - Clients that can't send a header, such as Claude Desktop, connect through the mcp-remote bridge, as Vexa's docs show. - Vexa is open source (Apache-2.0), so you can run it on your own servers too; a self-hosted install has fewer tools than the hosted service. Good to know - A real bot joins the call. By default it joins as a guest, so on Meet and Teams it can wait in the lobby until a host admits it, and the host can remove it. - On the hosted service, meetings are recorded by default (audio only) and meeting data is stored in the United States. Request the bot with recording_enabled set to false to transcribe without a recording. - Vexa's terms make you responsible for how it's used under your key, and its docs don't ask the other participants for consent, so tell people they're being transcribed and recorded wherever your law or company requires it. - speak_in_meeting talks out loud to everyone in the call and can't be taken back; Vexa says to use it only when the person it works for has asked. - Bot time is billed on the hosted service; new accounts start with free credit. - Type: Plugin - Made by: Vexa - Category: Communication - Formats: Remote MCP - Version: 0.12.27 - Website: https://docs.vexa.ai/vexa-mcp - Source: https://github.com/vexa-ai/vexa - Page: https://pluginboard.ai/plugins/vexa ## Works with - ChatGPT - Claude - Muse - Grok Bot - Gemini - Grok - GitHub Copilot - Cursor - Claude Code - Codex - Antigravity - Gemini CLI - Replit - OpenClaw - OpenCode - Hermes Agent ## Install You'll need: - `Authorization` (required): Bearer — from https://vexa.ai/account Replace each `` placeholder with your own value. ### ChatGPT In ChatGPT on the web: Plugins → + → Add custom MCP server, then paste https://api.cloud.vexa.ai/mcp under Connection. Business and Enterprise plans; Pro can add read-only servers. ### Claude In Claude (web or desktop): Customize → Connectors → Add custom connector, then paste https://api.cloud.vexa.ai/mcp ### Muse In Muse: In a Muse chat, ask it to create a custom connector for the MCP server at https://api.cloud.vexa.ai/mcp. Then tap Connect on the card it shows and sign in, or add your key. ### Grok Bot In Grok Bot: In a chat with your Bot, say "Add this MCP server: https://api.cloud.vexa.ai/mcp". It confirms the details with you and sets it up, and the tools are ready from your next message. ### Grok In Grok on the web: At grok.com/connectors: New Connector → Custom, then enter https://api.cloud.vexa.ai/mcp and sign in if it asks. ### GitHub Copilot Add to VS Code: vscode:mcp/install?%7B%22name%22%3A%22vexa%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fapi.cloud.vexa.ai%2Fmcp%22%7D Copilot CLI: ```shell copilot mcp add --transport http --header 'Authorization: ' vexa https://api.cloud.vexa.ai/mcp ``` ### Cursor Add to Cursor: cursor://anysphere.cursor-deeplink/mcp/install?name=vexa&config=eyJ1cmwiOiJodHRwczovL2FwaS5jbG91ZC52ZXhhLmFpL21jcCIsImhlYWRlcnMiOnsiQXV0aG9yaXphdGlvbiI6IjxBdXRob3JpemF0aW9uPiJ9fQ%3D%3D Add to .cursor/mcp.json: ```json { "mcpServers": { "vexa": { "url": "https://api.cloud.vexa.ai/mcp", "headers": { "Authorization": "" } } } } ``` ### Claude Code Run in your terminal: ```shell claude mcp add --transport http --header 'Authorization: ' vexa https://api.cloud.vexa.ai/mcp ``` ### Codex Run in your terminal: ```shell codex mcp add vexa --url https://api.cloud.vexa.ai/mcp ``` ### Antigravity Add to ~/.gemini/config/mcp_config.json: ```json { "mcpServers": { "vexa": { "url": "https://api.cloud.vexa.ai/mcp", "headers": { "Authorization": "" } } } } ``` ### Gemini CLI Run in your terminal: ```shell gemini mcp add --transport http --header 'Authorization: ' vexa https://api.cloud.vexa.ai/mcp ``` ### Replit In Replit: At replit.com/integrations, open MCP Servers → + Add MCP server, paste https://api.cloud.vexa.ai/mcp (add any headers under Advanced settings), then click Test & save. ### OpenClaw Run in your terminal: ```shell openclaw mcp add vexa --url https://api.cloud.vexa.ai/mcp --transport streamable-http ``` If it needs a key or sign-in: For https://api.cloud.vexa.ai/mcp, add --header NAME=value for each header it needs, or --auth oauth and then run openclaw mcp login vexa. ### OpenCode Run in your terminal: ```shell opencode mcp add vexa --url https://api.cloud.vexa.ai/mcp ``` If it needs a key: For https://api.cloud.vexa.ai/mcp, add --header 'NAME=value' for each header it needs. Add --global to use it in every project. ### Hermes Agent Run in your terminal: ```shell hermes mcp add vexa --url https://api.cloud.vexa.ai/mcp ``` Add to ~/.hermes/config.yaml: ```json { "mcp_servers": { "vexa": { "url": "https://api.cloud.vexa.ai/mcp", "headers": { "Authorization": "" } } } } ``` If it signs in with OAuth: If https://api.cloud.vexa.ai/mcp asks you to sign in, add --auth oauth to the command, or auth: oauth to its entry. --- From Pluginboard (https://pluginboard.ai).