Anthropic Tool Use (Function Calling)
Overview
Tool use lets Claude call functions you define — query databases, hit APIs, read files, do math. Claude decides when to call a tool, you execute it, and feed the result back. This is how you build Claude-powered agents.
Note: Anthropic does not offer an embeddings API. For embeddings + vector search, pair Claude with a dedicated embedding model (OpenAI, Cohere, or Voyage).
Prerequisites
- Completed
clade-model-inference
- Understanding of JSON Schema for tool definitions
Instructions
Step 1: Define Tools
import Anthropic from '@claude-ai/sdk';
const client = new Anthropic();
const tools: Anthropic.Tool[] = [
{
name: 'get_weather',
description: 'Get current weather for a city. Call this when the user asks about weather.',
input_schema: {
type: 'object',
properties: {
city: { type: 'string', description: 'City name, e.g. "San Francisco"' },
unit: { type: 'string', enum: ['celsius', 'fahrenheit'], description: 'Temperature unit' },
},
required: ['city'],
},
},
];
Step 2: Send Message with Tools
const response = await client.messages.create({
model: 'claude-sonnet-4-20250514',
max_tokens: 1024,
tools,
messages: [{ role: 'user', content: "What's the weather in San Francisco?" }],
});
Step 3: Execute Tool and Return Result
const toolUse = response.content.find(block => block.type === 'tool_use');
const weatherData = await fetchWeather(toolUse.input.city);
const finalResponse = await client.messages.create({
model: 'claude-sonnet-4-20250514',
max_tokens: 1024,
tools,
messages: [
{ role: 'user', content: "What's the weather in San Francisco?" },
{ role: 'assistant', content: response.content },
{
role: 'user',
content: [{
type: 'tool_result',
tool_use_id: toolUse.id,
content: JSON.stringify(weatherData),
}],
},
],
});
console.log(finalResponse.content[0].text);
Python Example
import anthropic
client = anthropic.Anthropic()
tools = [{
"name": "get_weather",
"description": "Get current weather for a city.",
"input_schema": {
"type": "object",
"properties": {
"city": {"type": "string"},
},
"required": ["city"],
},
}]
response = client.messages.create(
model="claude-sonnet-4-20250514",
max_tokens=1024,
tools=tools,
messages=[{"role": "user", "content": "Weather in Paris?"}],
)
for block in response.content:
if block.type == "tool_use":
result = execute_tool(block.name, block.input)
Agentic Tool Loop
let messages = [{ role: 'user', content: userInput }];
while (true) {
const response = await client.messages.create({
model: 'claude-sonnet-4-20250514',
max_tokens: 4096,
tools,
messages,
});
messages.push({ role: 'assistant', content: response.content });
if (response.stop_reason === 'end_turn') {
const text = response.content.find(b => b.type === 'text')?.text;
console.log(text);
break;
}
const toolResults = [];
for (const block of response.content) {
if (block.type === 'tool_use') {
const result = await executeTool(block.name, block.);
toolResults.({
: ,
: block.,
: .(result),
});
}
}
messages.({ : , : toolResults });
}
Output
tool_use content blocks with name and input when Claude wants to call a toolstop_reason: "tool_use" indicating Claude is waiting for tool results- Final text response after all tool results are provided
- Complete agentic loop until
stop_reason: "end_turn"
Error Handling
| Error | Cause | Solution |
|---|
invalid_request_error | Bad tool schema | Validate JSON Schema. input_schema must be a valid JSON Schema object |
tool_use with no matching name | Claude hallucinated a tool | Check tool_use.name against your defined tools before executing |
tool_result mismatch | Wrong tool_use_id | Each tool_result must reference the exact id from the tool_use block |
Examples
See Step 1 (tool definition), Step 2 (sending with tools), Step 3 (executing and returning results), and the full agentic tool loop example above.
Resources
Next Steps
See clade-common-errors for error handling patterns.
