Skip to content

Installation

This guide covers installing Syntha and setting up your development environment.

Requirements

  • Python 3.9 or higher
  • pip (Python package installer)

Basic Installation

Install Syntha using pip:

pip install syntha

That's it! Syntha works out of the box with SQLite (no additional setup required).

Verify Installation

Test your installation:

from syntha import ContextMesh, ToolHandler

# Create a test context mesh
context = ContextMesh(user_id="test_user")
handler = ToolHandler(context, "TestAgent")

print("✅ Syntha installed successfully!")

Development Installation

For development or contributing to Syntha:

# Clone the repository
git clone https://github.com/syntha/syntha-sdk.git
cd syntha-sdk

# Install in development mode
pip install -e .

# Install development dependencies
pip install -r requirements-test.txt

Database Backends

SQLite (Default)

No additional setup required. Great for: - Development and testing - Small teams (up to 100 users) - Single-server deployments

context = ContextMesh(
    user_id="user123",
    db_backend="sqlite",
    db_path="my_app.db"  # Optional: defaults to syntha.db
)

PostgreSQL (Production)

For production deployments with high concurrency:

# Install PostgreSQL driver
pip install psycopg2-binary

context = ContextMesh(
    user_id="user123",
    db_backend="postgresql",
    connection_string="postgresql://user:pass@localhost:5432/syntha_db"
)

Or use environment variables:

export CONNECTION_STRING="postgresql://user:pass@localhost:5432/syntha_db"

import os
context = ContextMesh(
    user_id="user123",
    db_backend="postgresql",
    connection_string=os.getenv("CONNECTION_STRING")
)

Framework Integration

Install additional packages for your LLM framework:

OpenAI

pip install openai

Anthropic Claude

pip install anthropic

LangChain

pip install langchain langchain-openai

Agno

pip install agno

Configuration

Environment Variables

Set these environment variables for your setup:

# Database (if using PostgreSQL)
export CONNECTION_STRING="postgresql://user:pass@localhost:5432/syntha_db"

# LLM API Keys
export OPENAI_API_KEY="your-openai-key"
export ANTHROPIC_API_KEY="your-anthropic-key"

# Optional: Logging level
export SYNTHA_LOG_LEVEL="INFO"

Configuration File

Create a syntha_config.py file:

# syntha_config.py
SYNTHA_CONFIG = {
    "db_backend": "postgresql",
    "connection_string": "postgresql://user:pass@localhost:5432/syntha_db",
    "enable_logging": True,
    "log_level": "INFO"
}

Docker Setup

Use Syntha with Docker:

FROM python:3.9-slim

WORKDIR /app

# Install Syntha
RUN pip install syntha

# For PostgreSQL support
RUN pip install psycopg2-binary

COPY . .

CMD ["python", "your_app.py"]

With docker-compose:

version: '3.8'
services:
  app:
    build: .
    environment:
      - DATABASE_URL=postgresql://user:pass@db:5432/syntha_db
      - OPENAI_API_KEY=${OPENAI_API_KEY}
    depends_on:
      - db

  db:
    image: postgres:13
    environment:
      - POSTGRES_DB=syntha_db
      - POSTGRES_USER=user
      - POSTGRES_PASSWORD=pass
    volumes:
      - postgres_data:/var/lib/postgresql/data

volumes:
  postgres_data:

Troubleshooting

Import Errors

If you see import errors:

# Ensure you're in the right Python environment
python --version
pip list | grep syntha

# Reinstall if needed
pip uninstall syntha
pip install syntha

Database Connection Issues

For PostgreSQL connection problems:

# Test connection
import psycopg2
conn = psycopg2.connect("postgresql://user:pass@localhost:5432/syntha_db")
print("✅ PostgreSQL connection successful")

Permission Issues

If you encounter permission errors:

# Install for current user only
pip install --user syntha

# Or use virtual environment
python -m venv syntha_env
source syntha_env/bin/activate  # On Windows: syntha_env\Scripts\activate
pip install syntha

Next Steps

Need Help?