Anthropic Integration
Memori integrates with all Anthropic Claude models, automatically capturing conversations through the Messages API to build persistent memory for your AI applications.
Installation
pip install memori anthropic
Quick Start
from anthropic import Anthropic
from memori import Memori
client = Anthropic()
# Register the Anthropic client with Memori
mem = Memori().llm.register(client)
mem.attribution(entity_id="user_123", process_id="claude_assistant")
response = client.messages.create(
model="claude-sonnet-4-5-20250929",
max_tokens=1024,
messages=[{"role": "user", "content": "Hello! My name is Alice."}]
)
print(response.content[0].text)
System Prompts
Anthropic uses a dedicated system parameter. Memori captures both system instructions and message content.
from anthropic import Anthropic
from memori import Memori
client = Anthropic()
mem = Memori().llm.register(client)
mem.attribution(entity_id="dev_001", process_id="coding_assistant")
response = client.messages.create(
model="claude-sonnet-4-5-20250929",
max_tokens=2048,
system="You are an expert Python developer. Provide concise, production-ready code.",
messages=[
{"role": "user", "content": "Write a function to validate email addresses"}
]
)
print(response.content[0].text)
Multi-Turn Conversations
Memori tracks conversation history across multiple API calls:
from anthropic import Anthropic
from memori import Memori
client = Anthropic()
mem = Memori().llm.register(client)
mem.attribution(entity_id="user_789", process_id="support")
messages = [
{"role": "user", "content": "I'm having trouble with my account password."}
]
# First turn
response = client.messages.create(
model="claude-sonnet-4-5-20250929",
max_tokens=1024,
messages=messages
)
messages.append({
"role": "assistant",
"content": response.content[0].text
})
# Second turn - memory is maintained
messages.append({
"role": "user",
"content": "Yes, I'd like to reset it."
})
response = client.messages.create(
model="claude-sonnet-4-5-20250929",
max_tokens=1024,
messages=messages
)
print(response.content[0].text)
from anthropic import Anthropic
from memori import Memori
import json
client = Anthropic()
mem = Memori().llm.register(client)
mem.attribution(entity_id="user_123", process_id="tools_demo")
tools = [
{
"name": "get_weather",
"description": "Get the current weather in a given location",
"input_schema": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
}
},
"required": ["location"]
}
}
]
response = client.messages.create(
model="claude-sonnet-4-5-20250929",
max_tokens=1024,
tools=tools,
messages=[{"role": "user", "content": "What's the weather in Paris?"}]
)
for content in response.content:
if content.type == "tool_use":
print(f"Tool: {content.name}")
print(f"Input: {json.dumps(content.input)}")
Beta Features
Memori supports Claude’s beta features including prompt caching and extended thinking:
from anthropic import Anthropic
from memori import Memori
client = Anthropic()
mem = Memori().llm.register(client)
mem.attribution(entity_id="user_123", process_id="beta_features")
# Using beta features
response = client.beta.messages.create(
model="claude-sonnet-4-5-20250929",
max_tokens=2048,
messages=[{"role": "user", "content": "Solve this complex problem..."}]
)
print(response.content[0].text)
Supported Features
| Feature | Support | Method |
|---|
| Sync Client | ✓ | Anthropic() |
| Async Client | ✓ | AsyncAnthropic() |
| Streaming | ✓ | client.messages.stream() |
| System Prompts | ✓ | system parameter |
| Tool Use | ✓ | tools parameter |
| Vision | ✓ | Multi-modal content |
| Prompt Caching | ✓ | Beta features |
| Beta API | ✓ | client.beta.messages |
How It Works
When you register an Anthropic client with Memori:
- Memori wraps
client.messages.create() and client.beta.messages.create()
- All message requests and responses are captured
- Conversations are stored in your Memori memory store
- A knowledge graph is built from conversation patterns
- The original API behavior is preserved
The max_tokens parameter is required by Anthropic’s API. Memori captures this along with all other request parameters.
Model Support
Memori works with all Claude models:
- Claude Sonnet 4 (claude-sonnet-4-5-20250929)
- Claude Sonnet 3.5 (claude-3-5-sonnet-20241022)
- Claude Haiku (claude-3-5-haiku-20241022)
- Claude Opus (claude-3-opus-20240229)
Next Steps