MS Teams AI BotInstall

Summary

Generate an AI-powered response in Microsoft Teams, optionally using conversation memory, and post it back to the channel.

Inputs

• aiProvider — AI provider to use (openai or anthropic).
• openaiApiKey — OpenAI API key (shown only when AI Provider is OpenAI).
• openaiModel — OpenAI model to use (e.g., GPT-4 Turbo).
• anthropicApiKey — Anthropic API key (shown only when AI Provider is Anthropic).
• anthropicModel — Anthropic model to use (e.g., Claude 3 Opus).
• systemPrompt — System prompt prepended to the conversation.
• enableMemory — Whether to retain conversation history for memory.
• memoryLimit — Maximum number of exchanges to keep in memory.
• options — Configuration options for the AI model (temperature, maxTokens, autoReply).

Output shape

array of objects each containing userMessage, aiResponse, conversationId, and memorySize

Returns a list of items; errors are thrown unless continueOnFail is enabled.

Examples

Example 1: Add a chat message from a Teams user and get an AI reply.

aiProvider=openai, openaiApiKey=..., openaiModel=gpt-4-turbo-preview, systemPrompt='You are a helpful AI assistant in Microsoft Teams. Be concise and professional.', enableMemory=true, memoryLimit=10, options={"temperature":0.7,"maxTokens":500,"autoReply":true}

Example 2: Process a user message with the Anthropic model and auto‑reply disabled.

aiProvider=anthropic, anthropicApiKey=..., anthropicModel=claude-3-sonnet-20240229, systemPrompt='You are a helpful AI assistant in Microsoft Teams. Be concise and professional.', enableMemory=false, options={"temperature":0.5,"maxTokens":300,"autoReply":false}