Quickstart

Add persistent memory to your AI agent in 5 minutes.

Full documentation: Introduction · REST API · MCP

1

Install the SDK

Install directly from GitHub. PyPI publishing coming soon.

bash
pip install "git+https://github.com/anonalabs/Anona-Memory-SDK.git"

For LiteLLM auto-injection also install: pip install litellm

2

Get your API key and Space ID

Create an API key on the API Keys page and a memory space on the Memory Spaces page.

API Keyanona_live_YOUR_KEY
Space IDYOUR_SPACE_ID
Gateway URLhttp://anona-prod-alb-747552680.us-east-1.elb.amazonaws.com
3

Add memory to your LLM calls

Option A: LiteLLM callback(recommended). Two lines. Every LLM call automatically retrieves relevant memories and injects them. After each response the Q&A turn is stored automatically.

python
import litellm
from anona.integrations.litellm import AnonaMemory

mem = AnonaMemory(
    api_key="anona_live_YOUR_KEY",
    space_id="YOUR_SPACE_ID",
    base_url="http://anona-prod-alb-747552680.us-east-1.elb.amazonaws.com",
)
mem.enable()  # registers callback with litellm

# All subsequent calls auto-inject memories + auto-store responses
response = litellm.completion(
    model="gemini/gemini-2.5-flash",  # any litellm model
    messages=[{"role": "user", "content": "What should I focus on today?"}],
)
print(response.choices[0].message.content)

Option B: Direct SDK. Full control: you decide what to store and when to recall.

python
from anona import AnonaClient

client = AnonaClient(
    api_key="anona_live_YOUR_KEY",
    base_url="http://anona-prod-alb-747552680.us-east-1.elb.amazonaws.com",
)

# Store a memory
client.add_memory(
    space_id="YOUR_SPACE_ID",
    content="User prefers concise Python answers.",
)

# Search memories (semantic)
results = client.search(
    space_id="YOUR_SPACE_ID",
    query="what does the user prefer?",
    limit=5,
)
for r in results:
    print(r["content"])

# Synthesise insights
summary = client.insights(
    space_id="YOUR_SPACE_ID",
    query="Summarise what you know about this user",
)
print(summary)

client.close()
4

Manual injection (any LLM / framework)

Not using LiteLLM? Retrieve memories and inject them yourself before calling any LLM.

python
from anona import AnonaClient

client = AnonaClient(api_key="anona_live_YOUR_KEY", base_url="http://anona-prod-alb-747552680.us-east-1.elb.amazonaws.com")

user_message = "What should I work on today?"

# 1. Retrieve relevant memories
memories = client.search(space_id="YOUR_SPACE_ID", query=user_message, limit=5)
memory_block = "\n".join(f"{i+1}. {r['content']}" for i, r in enumerate(memories))

# 2. Inject into messages
messages = []
if memory_block:
    messages.append({
        "role": "system",
        "content": f"# Relevant Memories\n{memory_block}",
    })
messages.append({"role": "user", "content": user_message})

# 3. Send to any LLM: OpenAI, Anthropic, Gemini, etc.
# response = your_llm_client.chat(messages)

# 4. Store the exchange
# client.add_memory(space_id="YOUR_SPACE_ID", content=f"User: {user_message}\nAssistant: {response}")

What's next