Skip to content
All Skills

Typespec Create Agent

Generate a complete TypeSpec declarative agent with instructions, capabilities, and conversation starters for Microsoft 365 Copilot

Agent Orchestration|v1|Updated 7/14/2026|GitHub source
MCP get_skill({ skillId: "create-typespec-declarative-agent-7b204709" })

Use this skill with your agent

Create a free account and connect via MCP

Get Started Free
# Create TypeSpec Declarative Agent

Create a complete TypeSpec declarative agent for Microsoft 365 Copilot with the following structure:

## Requirements

Generate a `main.tsp` file with:

1. **Agent Declaration**
   - Use `@agent` decorator with a descriptive name and description
   - Name should be 100 characters or less
   - Description should be 1,000 characters or less

2. **Instructions**
   - Use `@instructions` decorator with clear behavioral guidelines
   - Define the agent's role, expertise, and personality
   - Specify what the agent should and shouldn't do
   - Keep under 8,000 characters

3. **Conversation Starters**
   - Include 2-4 `@conversationStarter` decorators
   - Each with a title and example query
   - Make them diverse and showcase different capabilities

4. **Capabilities** (based on user needs)
   - `WebSearch` - for web content with optional site scoping
   - `OneDriveAndSharePoint` - for document access with URL filtering
   - `TeamsMessages` - for Teams channel/chat access
   - `Email` - for email access with folder filtering
   - `People` - for organization people search
   - `CodeInterpreter` - for Python code execution
   - `GraphicArt` - for image generation
   - `GraphConnectors` - for Copilot connector content
   - `Dataverse` - for Dataverse data access
   - `Meetings` - for meeting content access

## Template Structure

```typescript
import "@typespec/http";
import "@typespec/openapi3";
import "@microsoft/typespec-m365-copilot";

using TypeSpec.Http;
using TypeSpec.M365.Copilot.Agents;

@agent({
  name: "[Agent Name]",
  description: "[Agent Description]"
})
@instructions("""
  [Detailed instructions about agent behavior, role, and guidelines]
""")
@conversationStarter(#{
  title: "[Starter Title 1]",
  text: "[Example query 1]"
})
@conversationStarter(#{
  title: "[Starter Title 2]",
  text: "[Example query 2]"
})
namespace [AgentName] {
  // Add capabilities as operations here
  op capabilityName is AgentCapabilities.[CapabilityType]<[Parameters]>;
}
```

## Best Practices

- Use descriptive, role-based agent names (e.g., "Customer Support Assistant", "Research Helper")
- Write instructions in second person ("You are...")
- Be specific about the agent's expertise and limitations
- Include diverse conversation starters that showcase different features
- Only include capabilities the agent actually needs
- Scope capabilities (URLs, folders, etc.) when possible for better performance
- Use triple-quoted strings for multi-line instructions

## Examples

Ask the user:
1. What is the agent's purpose and role?
2. What capabilities does it need?
3. What knowledge sources should it access?
4. What are typical user interactions?

Then generate the complete TypeSpec agent definition.
#github-copilot#mcp#server

Related Skills

More skills in Agent Orchestration

1. Product type search — what design patterns fit this product?

Design UI/UX systems with style guides, palettes, typography, and component specs for new interfaces

#github#broad-capabilityMIT

Acp Router

Route plain-language requests for Claude Code, Cursor, Copilot, OpenClaw ACP, OpenCode, Gemini CLI, Qwen, Kiro, Kimi, iFlow, Factory Droid, Kilocode, or explicit ACP harness work into either OpenClaw ACP runtime sessions or direct acpx-driven sessions ("telephone game" flow). For coding-agent thread requests, read this skill first, then use only `sessions_spawn` for thread creation. Codex chat binding defaults to the native Codex app-server plugin unless ACP is explicit or background spawn needs ACP.

#broad-capability#browserMIT

Add New Safe Output Type

Adding a New Safe Output Type to GitHub Agentic Workflows

#github#broad-capabilityMIT

Agent Governance

Patterns and techniques for adding governance, safety, and trust controls to AI agent systems. Use this skill when: - Building AI agents that call external tools (APIs, databases, file systems) - Implementing policy-based access controls for agent tool usage - Adding semantic intent classification to detect dangerous prompts - Creating trust scoring systems for multi-agent workflows - Building audit trails for agent actions and decisions - Enforcing rate limits, content filters, or tool restrictions on agents - Working with any agent framework (PydanticAI, CrewAI, OpenAI Agents, LangChain, AutoGen)

#github-copilot#agentMIT

Agentic Development

Build AI agents with Pydantic AI (Python) and Claude SDK (Node.js)

#claude-bootstrap#bootstrapMIT

Agentic Workflows

Route gh-aw workflow design/create/debug/upgrade requests to the right prompts.

#broad-capability#agentic-workflowsMIT