◆ Developer Center

Build with Kyros

SDK reference, CLI commands, proxy configuration, and packaging guides — everything you need to integrate Kyros into your AI stack.

Getting Started

Up and running in 60 seconds

Install the SDK, set your API key, and make your first memory call. No infrastructure required for local development.

1

Prerequisites

PythonRequired: 3.9+python --version
PostgreSQLRequired: 15+ with pgvectorpsql --version
RedisRequired: 7+redis-server --version
Node.jsRequired: 18+ (for TypeScript SDK)node --version
2

Clone and install

terminal
bash
# Clone the repository
git clone https://github.com/Kyros-494/kyros-ai.git
cd kyros-ai

# Install the Python SDK in editable mode
pip install -e sdks/python

# With all optional integrations (LangChain, CrewAI, etc.)
pip install -e sdks/python[all]
3

Start the server

terminal
bash
# Start the Kyros API server (defaults to localhost:8000)
cd kyros-ai
uvicorn server.main:app --reload --port 8000

# In a separate terminal, verify it's running
curl http://localhost:8000/health
# → {"status":"ok","version":"1.0.0"}
4

Your first memory call

quickstart.py
python
import kyros

# Initialize the client
client = kyros.Client(
 api_key="mk_live_default_dev_key_123456",
 base_url="http://localhost:8000"
)

# Store a memory
result = client.ingest(
 content="User's name is Alex. Prefers Python over TypeScript.",
 user_id="user_001",
 type="semantic",
 tags=["user-profile", "preferences"]
)
print(f"Stored → hash: {result.hash[:16]}...")

# Recall relevant memories
memories = client.recall(
 query="What does the user prefer for development?",
 user_id="user_001",
 top_k=3,
 min_weight=0.2
)

for mem in memories:
 print(f"[{mem.retention_weight:.2f}] {mem.content}")
i
The default dev API key is mk_live_default_dev_key_123456. For production, generate a unique key using kyros tenant-create.
What's next?