Промпты в MCP-серверах

Введение в Model Context Protocol (MCP)

James Chapman

AI Curriculum Manager, DataCamp

Что такое MCP-промпты?

 

Промпты: многократно используемые шаблоны для оптимизации LLM под конкретные задачи

  • Рабочие процессы и поведение модели
  • Снижение нагрузки по составлению промптов для пользователей

prompts.png

Введение в Model Context Protocol (MCP)

Зачем нужны промпты

ambiguous_inputs.png

Введение в Model Context Protocol (MCP)

Промпт конвертера часовых поясов

You are a timezone conversion engine.

Your task is to:
1. Extract the source datetime from the user's natural language input.
2. Identify the source timezone (explicit or inferred).
3. Convert the datetime into the target timezone.

Rules:
- If the date is ambiguous (e.g., "next Friday"), resolve it relative to the provided datetime.
- If the input cannot be resolved confidently, seek clarification.

→ Опубликуйте этот промпт для задачи конвертации часовых поясов

Введение в Model Context Protocol (MCP)

Определение промптов в MCP-сервере

@mcp.prompt(title="Timezone Conversion")

def convert_timezone_prompt(user_input: str) -> str:
return f"""You are a timezone conversion engine. Your task is to: 1. Extract the source datetime... Rules: - If the date is ambiguous (e.g., "next Friday"), resolve it relative to the provided datetime. - If the input cannot be resolved confidently, seek clarification. User's timezone conversion request: {user_input}"""
Введение в Model Context Protocol (MCP)

Локальный MCP-сервер: timezone_server.py

from mcp.server.fastmcp import FastMCP

mcp = FastMCP("Timezone Converter")

# Tools and resources from before...

@mcp.prompt(title="Timezone Conversion")
def convert_timezone_prompt(timezone_request: str) -> str:
    # ...

if __name__ == "__main__":
    mcp.run(transport="stdio")
Введение в Model Context Protocol (MCP)

Клиент: получение списка промптов

async def list_prompts():
    """List all available prompts from the MCP server."""
    params = StdioServerParameters(command=sys.executable, args=["timezone_server.py"])

    async with stdio_client(params) as (reader, writer):
        async with ClientSession(reader, writer) as session:
            await session.initialize()

# List available prompts prompts = await session.list_prompts() print(f"Available prompts: {[p.name for p in prompts.prompts]}")
Введение в Model Context Protocol (MCP)

Клиент: получение списка промптов

print(asyncio.run(list_prompts()))
Available prompts: ['convert_timezone_prompt']

name промпта не совпадает с его title

@mcp.prompt(title="Timezone Conversion")
def convert_timezone_prompt(timezone_request: str) -> str:
    # ...
Введение в Model Context Protocol (MCP)

Клиент: получение списка промптов

print(asyncio.run(list_prompts()))
Available prompts: ['convert_timezone_prompt']

name промпта не совпадает с его title

# List available prompts
prompts = await session.list_prompts()
print(f"Available prompts: {[p.name for p in prompts.prompts]}")
  • На стороне клиента используется атрибут .name (не .title)
Введение в Model Context Protocol (MCP)

Клиент: получение промпта

async def read_prompt(user_input: str, prompt_name: str = "convert_timezone_prompt") -> str:
    """Retrieve a prompt from the MCP server with user input."""
    params = StdioServerParameters(command=sys.executable, args=["timezone_server.py"])

    async with stdio_client(params) as (reader, writer):
        async with ClientSession(reader, writer) as session:
            await session.initialize()

            prompts = await session.list_prompts()

# Retrieve the prompt if prompts.prompts: prompt = await session.get_prompt(prompt_name, arguments={"timezone_request": user_input})
print(f"Prompt result: {prompt.messages[0].content.text}") return prompt.messages[0].content.text
Введение в Model Context Protocol (MCP)
print(asyncio.run(read_prompt(user_input="It is 9:50 AM in the UK in January. What time is
    it in Lisbon, Portugal?")))
You are a timezone conversion engine.

Your task is to:
1. Extract the source datetime from the user's natural language input.
2. Identify the source timezone (explicit or inferred).
3. Convert the datetime into the target timezone.

Rules:
- If the date is ambiguous (e.g., "next Friday"), resolve it relative to the...
- If the input cannot be resolved confidently, seek clarification.

User's timezone conversion request: It is 9:50 AM in the UK in January. What time is it in
Lisbon, Portugal?
Введение в Model Context Protocol (MCP)

Вперёд, практиковаться!

Введение в Model Context Protocol (MCP)

Preparing Video For Download...