How to intermediate · 3 min read

How to use multiple tools with Claude API

Q: How to use multiple tools with Claude API

Use the anthropic Python SDK to call client.messages.create with system instructions that orchestrate multiple tools, or chain calls programmatically by invoking different tools sequentially. Manage tool integration logic in your code while leveraging claude-3-5-sonnet-20241022 for natural language understanding and control.

Quick answer

Use the anthropic Python SDK to call client.messages.create with system instructions that orchestrate multiple tools, or chain calls programmatically by invoking different tools sequentially. Manage tool integration logic in your code while leveraging claude-3-5-sonnet-20241022 for natural language understanding and control.

PREREQUISITES

Python 3.8+
Anthropic API key
pip install anthropic>=0.20

Setup

Install the Anthropic Python SDK and set your API key as an environment variable.

bash

pip install anthropic>=0.20

Step by step

This example demonstrates how to orchestrate multiple tools by calling the Claude API with system instructions and chaining calls in Python.

python

import os
import anthropic

client = anthropic.Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])

# Define system prompt to instruct Claude to use multiple tools
system_prompt = (
    "You are an assistant that uses two tools: a calculator and a dictionary. "
    "When given a math expression, use the calculator. When given a word, use the dictionary. "
    "Respond with the tool name and the result."
)

# User input that requires multiple tools
user_messages = [
    {"role": "user", "content": "Calculate 12 * 8 and define 'serendipity'."}
]

response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=300,
    system=system_prompt,
    messages=user_messages
)

print("Response from Claude:")
print(response.content[0].text)

# Alternatively, chain calls programmatically:

# 1. Calculator tool call
calc_prompt = "Calculate 12 * 8."
calc_response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=50,
    system="You are a calculator.",
    messages=[{"role": "user", "content": calc_prompt}]
)
calc_result = calc_response.content[0].text

# 2. Dictionary tool call
dict_prompt = "Define the word 'serendipity'."
dict_response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=100,
    system="You are a dictionary.",
    messages=[{"role": "user", "content": dict_prompt}]
)
dict_result = dict_response.content[0].text

print("\nChained tool results:")
print(f"Calculator result: {calc_result}")
print(f"Dictionary result: {dict_result}")

output

Response from Claude:
Calculator: 96
Dictionary: Serendipity means the occurrence of events by chance in a happy or beneficial way.

Chained tool results:
Calculator result: 96
Dictionary result: Serendipity means the occurrence of events by chance in a happy or beneficial way.

Common variations

You can use async calls with the Anthropic SDK or switch models like claude-3-opus-20240229 for different performance or cost profiles. Streaming responses are not currently supported in the Anthropic Python SDK.

python

import asyncio
import anthropic

async def async_example():
    client = anthropic.Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])
    response = await client.messages.acreate(
        model="claude-3-5-sonnet-20241022",
        max_tokens=100,
        system="You are a helpful assistant.",
        messages=[{"role": "user", "content": "Hello from async!"}]
    )
    print(response.content[0].text)

asyncio.run(async_example())

output

Hello from async!

Troubleshooting

If you get authentication errors, verify your ANTHROPIC_API_KEY environment variable is set correctly.
If responses are incomplete, increase max_tokens.
For unexpected behavior, ensure your system prompt clearly defines tool usage instructions.

✅

Key Takeaways

Use the system parameter to instruct Claude on how to use multiple tools in one call.
Chain multiple client.messages.create calls programmatically to integrate distinct tools.
Use the latest Claude model claude-3-5-sonnet-20241022 for best coding and reasoning performance.
Always set your API key via environment variables to avoid security risks.
Async calls are supported in the Anthropic SDK for scalable tool orchestration.

Verified 2026-04 · claude-3-5-sonnet-20241022, claude-3-opus-20240229

Verify ↗