Construire avec la réflexion étendue

weather_tool = {
    "name": "get_weather",
    "description": "Get current weather for a location",
    "input_schema": {
        "type": "object",
        "properties": {
            "location": {"type": "string"}
        },
        "required": ["location"]
    }
}

# First request - Claude responds with thinking and tool request
response = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    tools=[weather_tool],
    messages=[
        {"role": "user", "content": "What's the weather in Paris?"}
    ]
)

# Extract thinking block and tool use block
thinking_block = next((block for block in response.content
                      if block.type == 'thinking'), None)
tool_use_block = next((block for block in response.content
                      if block.type == 'tool_use'), None)

# Call your actual weather API, here is where your actual API call would go
# let's pretend this is what we get back
weather_data = {"temperature": 88}

# Second request - Include thinking block and tool result
# No new thinking blocks will be generated in the response
continuation = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    tools=[weather_tool],
    messages=[
        {"role": "user", "content": "What's the weather in Paris?"},
        # notice that the thinking_block is passed in as well as the tool_use_block
        # if this is not passed in, an error is raised
        {"role": "assistant", "content": [thinking_block, tool_use_block]},
        {"role": "user", "content": [{
            "type": "tool_result",
            "tool_use_id": tool_use_block.id,
            "content": f"Current temperature: {weather_data['temperature']}°F"
        }]}
    ]
)

import anthropic

client = anthropic.Anthropic()

# Define tools
calculator_tool = {
    "name": "calculator",
    "description": "Perform mathematical calculations",
    "input_schema": {
        "type": "object",
        "properties": {
            "expression": {
                "type": "string",
                "description": "Mathematical expression to evaluate"
            }
        },
        "required": ["expression"]
    }
}

database_tool = {
    "name": "database_query",
    "description": "Query product database",
    "input_schema": {
        "type": "object",
        "properties": {
            "query": {
                "type": "string",
                "description": "SQL query to execute"
            }
        },
        "required": ["query"]
    }
}

# First request - Claude thinks once before all tool calls
response = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    tools=[calculator_tool, database_tool],
    messages=[{
        "role": "user",
        "content": "What's the total revenue if we sold 150 units of product A at $50 each, and how does this compare to our average monthly revenue from the database?"
    }]
)

# Response includes thinking followed by tool uses
# Note: Claude thinks once at the beginning, then makes all tool decisions
print("First response:")
for block in response.content:
    if block.type == "thinking":
        print(f"Thinking (summarized): {block.thinking}")
    elif block.type == "tool_use":
        print(f"Tool use: {block.name} with input {block.input}")
    elif block.type == "text":
        print(f"Text: {block.text}")

# You would execute the tools and return results...
# After getting both tool results back, Claude directly responds without additional thinking

import anthropic

client = anthropic.Anthropic()

# Same tool definitions as before
calculator_tool = {
    "name": "calculator",
    "description": "Perform mathematical calculations",
    "input_schema": {
        "type": "object",
        "properties": {
            "expression": {
                "type": "string",
                "description": "Mathematical expression to evaluate"
            }
        },
        "required": ["expression"]
    }
}

database_tool = {
    "name": "database_query",
    "description": "Query product database",
    "input_schema": {
        "type": "object",
        "properties": {
            "query": {
                "type": "string",
                "description": "SQL query to execute"
            }
        },
        "required": ["query"]
    }
}

# First request with interleaved thinking enabled
response = client.beta.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    tools=[calculator_tool, database_tool],
    betas=["interleaved-thinking-2025-05-14"],
    messages=[{
        "role": "user",
        "content": "What's the total revenue if we sold 150 units of product A at $50 each, and how does this compare to our average monthly revenue from the database?"
    }]
)

print("Initial response:")
thinking_blocks = []
tool_use_blocks = []

for block in response.content:
    if block.type == "thinking":
        thinking_blocks.append(block)
        print(f"Thinking: {block.thinking}")
    elif block.type == "tool_use":
        tool_use_blocks.append(block)
        print(f"Tool use: {block.name} with input {block.input}")
    elif block.type == "text":
        print(f"Text: {block.text}")

# First tool result (calculator)
calculator_result = "7500"  # 150 * 50

# Continue with first tool result
response2 = client.beta.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    tools=[calculator_tool, database_tool],
    betas=["interleaved-thinking-2025-05-14"],
    messages=[
        {
            "role": "user",
            "content": "What's the total revenue if we sold 150 units of product A at $50 each, and how does this compare to our average monthly revenue from the database?"
        },
        {
            "role": "assistant",
            "content": [thinking_blocks[0], tool_use_blocks[0]]
        },
        {
            "role": "user",
            "content": [{
                "type": "tool_result",
                "tool_use_id": tool_use_blocks[0].id,
                "content": calculator_result
            }]
        }
    ]
)

print("\nAfter calculator result:")
# With interleaved thinking, Claude can think about the calculator result
# before deciding to query the database
for block in response2.content:
    if block.type == "thinking":
        thinking_blocks.append(block)
        print(f"Interleaved thinking: {block.thinking}")
    elif block.type == "tool_use":
        tool_use_blocks.append(block)
        print(f"Tool use: {block.name} with input {block.input}")

# Second tool result (database)
database_result = "5200"  # Example average monthly revenue

# Continue with second tool result
response3 = client.beta.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    tools=[calculator_tool, database_tool],
    betas=["interleaved-thinking-2025-05-14"],
    messages=[
        {
            "role": "user",
            "content": "What's the total revenue if we sold 150 units of product A at $50 each, and how does this compare to our average monthly revenue from the database?"
        },
        {
            "role": "assistant",
            "content": [thinking_blocks[0], tool_use_blocks[0]]
        },
        {
            "role": "user",
            "content": [{
                "type": "tool_result",
                "tool_use_id": tool_use_blocks[0].id,
                "content": calculator_result
            }]
        },
        {
            "role": "assistant",
            "content": thinking_blocks[1:] + tool_use_blocks[1:]
        },
        {
            "role": "user",
            "content": [{
                "type": "tool_result",
                "tool_use_id": tool_use_blocks[1].id,
                "content": database_result
            }]
        }
    ]
)

print("\nAfter database result:")
# With interleaved thinking, Claude can think about both results
# before formulating the final response
for block in response3.content:
    if block.type == "thinking":
        print(f"Final thinking: {block.thinking}")
    elif block.type == "text":
        print(f"Final response: {block.text}")

from anthropic import Anthropic
import requests
from bs4 import BeautifulSoup

client = Anthropic()

def fetch_article_content(url):
    response = requests.get(url)
    soup = BeautifulSoup(response.content, 'html.parser')

    # Remove script and style elements
    for script in soup(["script", "style"]):
        script.decompose()

    # Get text
    text = soup.get_text()

    # Break into lines and remove leading and trailing space on each
    lines = (line.strip() for line in text.splitlines())
    # Break multi-headlines into a line each
    chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
    # Drop blank lines
    text = '\n'.join(chunk for chunk in chunks if chunk)

    return text

# Fetch the content of the article
book_url = "https://www.gutenberg.org/cache/epub/1342/pg1342.txt"
book_content = fetch_article_content(book_url)
# Use just enough text for caching (first few chapters)
LARGE_TEXT = book_content[:5000]

SYSTEM_PROMPT=[
    {
        "type": "text",
        "text": "You are an AI assistant that is tasked with literary analysis. Analyze the following text carefully.",
    },
    {
        "type": "text",
        "text": LARGE_TEXT,
        "cache_control": {"type": "ephemeral"}
    }
]

MESSAGES = [
    {
        "role": "user",
        "content": "Analyze the tone of this passage."
    }
]

# First request - establish cache
print("First request - establishing cache")
response1 = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=20000,
    thinking={
        "type": "enabled",
        "budget_tokens": 4000
    },
    system=SYSTEM_PROMPT,
    messages=MESSAGES
)

print(f"First response usage: {response1.usage}")

MESSAGES.append({
    "role": "assistant",
    "content": response1.content
})
MESSAGES.append({
    "role": "user",
    "content": "Analyze the characters in this passage."
})
# Second request - same thinking parameters (cache hit expected)
print("\nSecond request - same thinking parameters (cache hit expected)")
response2 = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=20000,
    thinking={
        "type": "enabled",
        "budget_tokens": 4000
    },
    system=SYSTEM_PROMPT,
    messages=MESSAGES
)

print(f"Second response usage: {response2.usage}")

# Third request - different thinking parameters (cache miss for messages)
print("\nThird request - different thinking parameters (cache miss for messages)")
response3 = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=20000,
    thinking={
        "type": "enabled",
        "budget_tokens": 8000  # Changed thinking budget
    },
    system=SYSTEM_PROMPT,  # System prompt remains cached
    messages=MESSAGES  # Messages cache is invalidated
)

print(f"Third response usage: {response3.usage}")

from anthropic import Anthropic
import requests
from bs4 import BeautifulSoup

client = Anthropic()

def fetch_article_content(url):
    response = requests.get(url)
    soup = BeautifulSoup(response.content, 'html.parser')

    # Remove script and style elements
    for script in soup(["script", "style"]):
        script.decompose()

    # Get text
    text = soup.get_text()

    # Break into lines and remove leading and trailing space on each
    lines = (line.strip() for line in text.splitlines())
    # Break multi-headlines into a line each
    chunks = (phrase.strip() for line in lines for phrase in line.split("  "))
    # Drop blank lines
    text = '\n'.join(chunk for chunk in chunks if chunk)

    return text

# Fetch the content of the article
book_url = "https://www.gutenberg.org/cache/epub/1342/pg1342.txt"
book_content = fetch_article_content(book_url)
# Use just enough text for caching (first few chapters)
LARGE_TEXT = book_content[:5000]

# No system prompt - caching in messages instead
MESSAGES = [
    {
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": LARGE_TEXT,
                "cache_control": {"type": "ephemeral"},
            },
            {
                "type": "text",
                "text": "Analyze the tone of this passage."
            }
        ]
    }
]

# First request - establish cache
print("First request - establishing cache")
response1 = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=20000,
    thinking={
        "type": "enabled",
        "budget_tokens": 4000
    },
    messages=MESSAGES
)

print(f"First response usage: {response1.usage}")

MESSAGES.append({
    "role": "assistant",
    "content": response1.content
})
MESSAGES.append({
    "role": "user",
    "content": "Analyze the characters in this passage."
})
# Second request - same thinking parameters (cache hit expected)
print("\nSecond request - same thinking parameters (cache hit expected)")
response2 = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=20000,
    thinking={
        "type": "enabled",
        "budget_tokens": 4000  # Same thinking budget
    },
    messages=MESSAGES
)

print(f"Second response usage: {response2.usage}")

MESSAGES.append({
    "role": "assistant",
    "content": response2.content
})
MESSAGES.append({
    "role": "user",
    "content": "Analyze the setting in this passage."
})

# Third request - different thinking budget (cache miss expected)
print("\nThird request - different thinking budget (cache miss expected)")
response3 = client.messages.create(
    model="claude-sonnet-4-5",
    max_tokens=20000,
    thinking={
        "type": "enabled",
        "budget_tokens": 8000  # Different thinking budget breaks cache
    },
    messages=MESSAGES
)

print(f"Third response usage: {response3.usage}")

import anthropic

client = anthropic.Anthropic()

# Using a special prompt that triggers redacted thinking (for demonstration purposes only)
response = client.messages.create(
    model="claude-sonnet-4-5-20250929",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    messages=[{
        "role": "user",
        "content": "ANTHROPIC_MAGIC_STRING_TRIGGER_REDACTED_THINKING_46C9A13E193C177646C7398A98432ECCCE4C1253D5E2D82641AC0E52CC2876CB"
    }]
)

# Identify redacted thinking blocks
has_redacted_thinking = any(
    block.type == "redacted_thinking" for block in response.content
)

if has_redacted_thinking:
    print("Response contains redacted thinking blocks")
    # These blocks are still usable in subsequent requests

    # Extract all blocks (both redacted and non-redacted)
    all_thinking_blocks = [
        block for block in response.content
        if block.type in ["thinking", "redacted_thinking"]
    ]

    # When passing to subsequent requests, include all blocks without modification
    # This preserves the integrity of Claude's reasoning

    print(f"Found {len(all_thinking_blocks)} thinking blocks total")
    print(f"These blocks are still billable as output tokens")