Skip to main content

​
Configuration Guide

OmniCoreAgent can be configured through environment variables, Python dictionaries, or specialized configuration objects.

​
1. Environment Variables

Environment variables are the best way to manage sensitive data like API keys and connection strings.

​
LLM API Keys

ProviderVariable
OpenAIOPENAI_API_KEY
AnthropicANTHROPIC_API_KEY
Google GeminiGEMINI_API_KEY
GroqGROQ_API_KEY

​
Storage Backends

BackendVariableExample
RedisREDIS_URLredis://localhost:6379/0
SQL DatabaseDATABASE_URLpostgresql://user:pass@localhost:5432/db
MongoDBMONGODB_URImongodb://localhost:27017

​
Observability

ServiceVariable
OpikOPIK_API_KEY
Opik ProjectOPIK_PROJECT_NAME

​
2. Agent Configuration

The AgentConfig handles the runtime behavior of the agent, such as reasoning steps and resource limits. 
agent_config = {
    "tool_call_timeout": 30,         # Max seconds per tool execution
    "max_steps": 15,                 # Max reasoning loops per run()
    "request_limit": 1000,           # Max LLM requests per session
    "total_tokens_limit": 100000,    # Max total tokens per session
    "enable_agent_skills": True,     # Enable local skill discovery
    "enable_advanced_tool_use": True # Enable BM25 tool retrieval
}

agent = OmniCoreAgent(
    ...
    agent_config=agent_config
)

​
3. Model Configuration

The model_config defines which LLM to use and its sampling parameters. 
model_config = {
    "provider": "openai",
    "model": "gpt-4o",
    "temperature": 0.5,
    "max_tokens": 4000,
    "top_p": 0.9,
    # Custom endpoint for self-hosted models
    "api_base": "http://localhost:8000/v1" 
}

​
4. MCP Tool Configuration

MCP servers are configured as a list of dictionaries. 
mcp_tools = [
    {
        "name": "explorer",
        "transport_type": "stdio",
        "command": "npx",
        "args": ["-y", "@modelcontextprotocol/server-filesystem", "/tmp"]
    },
    {
        "name": "remote_service",
        "transport_type": "streamable_http",
        "url": "https://api.myapp.com/mcp",
        "headers": {"Authorization": "Bearer token"}
    }
]

​
5. Persistence Configuration

Pass a MemoryRouter or EventRouter to customize where history and telemetry are stored. 
from omnicoreagent import MemoryRouter, EventRouter

agent = OmniCoreAgent(
    ...
    memory_router=MemoryRouter("redis"),
    event_router=EventRouter("redis_stream")
)

​
Best Practices

  • Use .env: Use a library like python-dotenv to load your environment variables during development.
  • Model Selection: Use smaller models (gpt-4o-mini) while building and testing your agent logic to save costs.
  • Limit Steps: Always set a reasonable max_steps to prevent the agent from entering infinite reasoning loops in case of tool failures.