home / skills / sickn33 / antigravity-awesome-skills / m365-agents-ts

m365-agents-ts skill

/skills/m365-agents-ts

This skill helps you build Microsoft 365 agents in TypeScript with Express hosting, streaming, and Copilot Studio integration for multichannel conversations.

npx playbooks add skill sickn33/antigravity-awesome-skills --skill m365-agents-ts

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
6.3 KB
---
name: m365-agents-ts
description: |
  Microsoft 365 Agents SDK for TypeScript/Node.js. Build multichannel agents for Teams/M365/Copilot Studio with AgentApplication routing, Express hosting, streaming responses, and Copilot Studio client integration. Triggers: "Microsoft 365 Agents SDK", "@microsoft/agents-hosting", "AgentApplication", "startServer", "streamingResponse", "Copilot Studio client", "@microsoft/agents-copilotstudio-client".
package: "@microsoft/agents-hosting, @microsoft/agents-hosting-express, @microsoft/agents-activity, @microsoft/agents-copilotstudio-client"
---

# Microsoft 365 Agents SDK (TypeScript)

Build enterprise agents for Microsoft 365, Teams, and Copilot Studio using the Microsoft 365 Agents SDK with Express hosting, AgentApplication routing, streaming responses, and Copilot Studio client integrations.

## Before implementation
- Use the microsoft-docs MCP to verify the latest API signatures for AgentApplication, startServer, and CopilotStudioClient.
- Confirm package versions on npm before wiring up samples or templates.

## Installation

```bash
npm install @microsoft/agents-hosting @microsoft/agents-hosting-express @microsoft/agents-activity
npm install @microsoft/agents-copilotstudio-client
```

## Environment Variables

```bash
PORT=3978
AZURE_RESOURCE_NAME=<azure-openai-resource>
AZURE_API_KEY=<azure-openai-key>
AZURE_OPENAI_DEPLOYMENT_NAME=gpt-4o-mini

TENANT_ID=<tenant-id>
CLIENT_ID=<client-id>
CLIENT_SECRET=<client-secret>

COPILOT_ENVIRONMENT_ID=<environment-id>
COPILOT_SCHEMA_NAME=<schema-name>
COPILOT_CLIENT_ID=<copilot-app-client-id>
COPILOT_BEARER_TOKEN=<copilot-jwt>
```

## Core Workflow: Express-hosted AgentApplication

```typescript
import { AgentApplication, TurnContext, TurnState } from "@microsoft/agents-hosting";
import { startServer } from "@microsoft/agents-hosting-express";

const agent = new AgentApplication<TurnState>();

agent.onConversationUpdate("membersAdded", async (context: TurnContext) => {
  await context.sendActivity("Welcome to the agent.");
});

agent.onMessage("hello", async (context: TurnContext) => {
  await context.sendActivity(`Echo: ${context.activity.text}`);
});

startServer(agent);
```

## Streaming responses with Azure OpenAI

```typescript
import { azure } from "@ai-sdk/azure";
import { AgentApplication, TurnContext, TurnState } from "@microsoft/agents-hosting";
import { startServer } from "@microsoft/agents-hosting-express";
import { streamText } from "ai";

const agent = new AgentApplication<TurnState>();

agent.onMessage("poem", async (context: TurnContext) => {
  context.streamingResponse.setFeedbackLoop(true);
  context.streamingResponse.setGeneratedByAILabel(true);
  context.streamingResponse.setSensitivityLabel({
    type: "https://schema.org/Message",
    "@type": "CreativeWork",
    name: "Internal",
  });

  await context.streamingResponse.queueInformativeUpdate("starting a poem...");

  const { fullStream } = streamText({
    model: azure(process.env.AZURE_OPENAI_DEPLOYMENT_NAME || "gpt-4o-mini"),
    system: "You are a creative assistant.",
    prompt: "Write a poem about Apollo.",
  });

  try {
    for await (const part of fullStream) {
      if (part.type === "text-delta" && part.text.length > 0) {
        await context.streamingResponse.queueTextChunk(part.text);
      }
      if (part.type === "error") {
        throw new Error(`Streaming error: ${part.error}`);
      }
    }
  } finally {
    await context.streamingResponse.endStream();
  }
});

startServer(agent);
```

## Invoke activity handling

```typescript
import { Activity, ActivityTypes } from "@microsoft/agents-activity";
import { AgentApplication, TurnContext, TurnState } from "@microsoft/agents-hosting";

const agent = new AgentApplication<TurnState>();

agent.onActivity("invoke", async (context: TurnContext) => {
  const invokeResponse = Activity.fromObject({
    type: ActivityTypes.InvokeResponse,
    value: { status: 200 },
  });

  await context.sendActivity(invokeResponse);
  await context.sendActivity("Thanks for submitting your feedback.");
});
```

## Copilot Studio client (Direct to Engine)

```typescript
import { CopilotStudioClient } from "@microsoft/agents-copilotstudio-client";

const settings = {
  environmentId: process.env.COPILOT_ENVIRONMENT_ID!,
  schemaName: process.env.COPILOT_SCHEMA_NAME!,
  clientId: process.env.COPILOT_CLIENT_ID!,
};

const tokenProvider = async (): Promise<string> => {
  return process.env.COPILOT_BEARER_TOKEN!;
};

const client = new CopilotStudioClient(settings, tokenProvider);

const conversation = await client.startConversationAsync();
const reply = await client.askQuestionAsync("Hello!", conversation.id);
console.log(reply);
```

## Copilot Studio WebChat integration

```typescript
import { CopilotStudioWebChat } from "@microsoft/agents-copilotstudio-client";

const directLine = CopilotStudioWebChat.createConnection(client, {
  showTyping: true,
});

window.WebChat.renderWebChat({
  directLine,
}, document.getElementById("webchat")!);
```

## Best Practices

1. Use AgentApplication for routing and keep handlers focused on one responsibility.
2. Prefer streamingResponse for long-running completions and call endStream in finally blocks.
3. Keep secrets out of source code; load tokens from environment variables or secure stores.
4. Reuse CopilotStudioClient instances and cache tokens in your token provider.
5. Validate invoke payloads before logging or persisting feedback.

## Reference Files

| File | Contents |
| --- | --- |
| [references/acceptance-criteria.md](references/acceptance-criteria.md) | Import paths, hosting pipeline, streaming, and Copilot Studio patterns |

## Reference Links

| Resource | URL |
| --- | --- |
| Microsoft 365 Agents SDK | https://learn.microsoft.com/en-us/microsoft-365/agents-sdk/ |
| JavaScript SDK overview | https://learn.microsoft.com/en-us/javascript/api/overview/agents-overview?view=agents-sdk-js-latest |
| @microsoft/agents-hosting-express | https://learn.microsoft.com/en-us/javascript/api/%40microsoft/agents-hosting-express?view=agents-sdk-js-latest |
| @microsoft/agents-copilotstudio-client | https://learn.microsoft.com/en-us/javascript/api/%40microsoft/agents-copilotstudio-client?view=agents-sdk-js-latest |
| Integrate with Copilot Studio | https://learn.microsoft.com/en-us/microsoft-365/agents-sdk/integrate-with-mcs |
| GitHub samples | https://github.com/microsoft/Agents/tree/main/samples/nodejs |

Overview

This skill provides a TypeScript/Node.js implementation of the Microsoft 365 Agents SDK for building multichannel agents targeting Teams, Microsoft 365, and Copilot Studio. It includes patterns for AgentApplication routing, Express hosting, streaming responses with Azure OpenAI, and direct Copilot Studio client integrations. The samples focus on production-ready practices like token handling, streaming feedback loops, and invoke activity handling.

How this skill works

The skill wires an AgentApplication to an Express host and registers handlers for conversation updates, message triggers, and invoke activities. It demonstrates streaming completions by piping Azure OpenAI stream parts into the SDK streamingResponse API, including feedback loop and sensitivity labels. It also shows how to authenticate and reuse a CopilotStudioClient to start conversations and connect WebChat via a DirectLine-like connection.

When to use it

  • Building Teams or M365 chat agents that require structured routing and lifecycle hooks
  • Implementing real-time streaming responses for long-running model outputs
  • Connecting server-side logic directly to Copilot Studio engines or WebChat frontends
  • Handling invoke activities or custom payloads from Teams and other M365 surfaces
  • Prototyping production agents where token reuse and secure environment config are required

Best practices

  • Use AgentApplication to keep handlers small and single-responsibility
  • Prefer streamingResponse for long outputs and always call endStream in a finally block
  • Keep secrets out of source: load keys and tokens from environment variables or secure stores
  • Reuse CopilotStudioClient instances and implement token caching in the provider
  • Validate and sanitize invoke payloads before logging or persisting feedback

Example use cases

  • A Teams onboarding bot that welcomes new members and routes commands via AgentApplication
  • A creative assistant that streams poem generation from Azure OpenAI into the chat UI using streamingResponse
  • A feedback form that handles invoke activities and responds with an InvokeResponse then stores user input
  • A Copilot Studio integration that starts a conversation server-side and exposes a WebChat front-end via CopilotStudioWebChat
  • A multitenant M365 assistant that uses environment-based credentials and cached Copilot tokens

FAQ

Which environment variables are required?

Set host port plus Azure OpenAI and Azure AD/Copilot secrets such as AZURE_RESOURCE_NAME, AZURE_API_KEY, AZURE_OPENAI_DEPLOYMENT_NAME, TENANT_ID, CLIENT_ID, CLIENT_SECRET, and COPILOT_* values.

How do I handle streaming errors?

Listen for error parts from the model stream, throw or log as needed, and ensure streamingResponse.endStream() is called in a finally block to gracefully terminate the client stream.