StandIn

StandIn is an AI agent tool that puts your own AI agent into real Microsoft Teams calls as a participant, aimed at developers and teams who want an AI teammate that listens, sees the screen, speaks, and acts. You connect your own agent over a WebSocket, so StandIn acts as the Teams bridge rather than the model, and you keep full control of the AI. In a call it becomes a real participant: live voice, screen-share vision, a lip-synced avatar tile, and actions, responding when it is addressed. StandIn is built around a sovereign, bring-your-own-Azure-AD model, so the bot identity and all meeting content stay in your own Microsoft tenant; the service transits media but stores none, meaning your keys, prompts, and transcripts never leave your infrastructure. You can try it in seconds in a free sandbox with a quick Microsoft or Google sign-in, with no Azure Bot, no admin consent, and no card, then deploy to your own tenant through a BYO-AAD wizard that guides Azure Bot setup and generates a Teams app package. The Teams Voice plugin is open source and runs on your own infrastructure. It is best for teams that need an AI Teams participant while keeping data inside their tenant.

Best for developers and teams that want their own AI agent to join Microsoft Teams calls while keeping data inside their tenant.