System prompt caching pattern

import anthropic
import os

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

# System prompt to cache: must be identical across requests to reuse cache
system_prompt = """You are an expert legal document reviewer. Your task is to:
1. Identify contract risks and red flags
2. Highlight missing clauses or ambiguous language
3. Provide specific remediation suggestions
4. Rate overall risk (low/medium/high)

Always cite the specific clause or section when flagging issues.
Be concise but thorough. Assume the reader has legal background."""

# First request: cache miss, full processing
response_1 = client.messages.create(
    model="claude-opus-4-6",
    max_tokens=1024,
    system=[
        {
            "type": "text",
            "text": system_prompt,
            "cache_control": {"type": "ephemeral"}
        }
    ],
    messages=[
        {
            "role": "user",
            "content": "Review this NDA: [contract text here]"
        }
    ]
)

print(f"First request usage:")
print(f"Input tokens: {response_1.usage.input_tokens}")
print(f"Cache creation tokens: {response_1.usage.cache_creation_input_tokens}")
print(f"Cache read tokens: {response_1.usage.cache_read_input_tokens}")
print(f"Response: {response_1.content[0].text}\n")

# Second request: same system prompt within 5 min, should hit cache
response_2 = client.messages.create(
    model="claude-opus-4-6",
    max_tokens=1024,
    system=[
        {
            "type": "text",
            "text": system_prompt,
            "cache_control": {"type": "ephemeral"}
        }
    ],
    messages=[
        {
            "role": "user",
            "content": "Review this employment agreement: [different contract]"
        }
    ]
)

print(f"Second request usage (should show cache_read_input_tokens):")
print(f"Input tokens: {response_2.usage.input_tokens}")
print(f"Cache creation tokens: {response_2.usage.cache_creation_input_tokens}")
print(f"Cache read tokens: {response_2.usage.cache_read_input_tokens}")
print(f"Response: {response_2.content[0].text}")