Skip to main content

Configuration Overview

AgentMap provides a sophisticated configuration system that supports multiple deployment scenarios, advanced LLM routing, comprehensive storage backends, and extensive customization options. This guide covers the complete configuration capabilities available in AgentMap.

🏗️ Configuration Architecture​

AgentMap uses a hierarchical configuration system with multiple layers:

  1. Main Configuration File (agentmap_config.yaml) - Core application settings
  2. Storage Configuration File (agentmap_config_storage.yaml) - Storage provider settings
  3. Environment Variables - Runtime overrides and sensitive credentials
  4. Default Values - Built-in fallbacks for all optional settings

📋 Quick Configuration Checklist​

Basic Setup (5 minutes)​

Production Setup (30 minutes)​

Advanced Setup (1 hour)​

🚀 Configuration Capabilities​

Core Features​

  • Multiple LLM Providers: OpenAI, Anthropic, Google with automatic routing
  • Storage Backends: CSV, JSON, Vector databases, Cloud storage, Firebase
  • Authentication: JWT, API keys, Supabase integration
  • Environment Overrides: Complete runtime configuration control

Advanced Features​

  • Intelligent LLM Routing: Provider × complexity matrix with cost optimization
  • Multi-Provider Storage: Seamless switching between local and cloud storage
  • Memory Management: Multiple buffer types with configurable policies
  • Execution Tracking: Comprehensive workflow monitoring and success policies
  • Performance Optimization: Caching, cost controls, and resource management

Enterprise Features​

  • Service Discovery: Host application protocol integration
  • Tracing & Monitoring: LangSmith integration with local export options
  • Security: Comprehensive authentication and authorization system
  • Scalability: Configuration for high-throughput production deployments

📚 Configuration Guide Structure​

Main Configuration​

Complete YAML structure covering all core AgentMap settings including LLM providers, routing, memory, execution, and performance optimization.

Storage Configuration​

Comprehensive guide to storage providers including CSV, JSON, Vector databases, Firebase, and cloud storage with authentication and optimization settings.

Environment Variables​

Complete reference of environment variables for API keys, storage credentials, system overrides, and security settings.

Configuration Examples​

Working configuration examples for different deployment scenarios including local development, production, and cloud integration.

Troubleshooting​

Common configuration issues, validation errors, and debugging guidance with solutions.

🎯 Configuration by Use Case​

Local Development​

# Minimal configuration for getting started
csv_path: "examples/HelloWorld.csv"
autocompile: true
llm:
  openai:
    api_key: "env:OPENAI_API_KEY"
    model: "gpt-3.5-turbo"

Production Deployment​

# Production-ready configuration with routing and optimization
csv_path: "workflows/production.csv"
autocompile: true
storage_config_path: "agentmap_config_storage.yaml"

routing:
  enabled: true
  cost_optimization:
    enabled: true
    prefer_cost_effective: true
    max_cost_tier: "high"

logging:
  file_path: "/var/log/agentmap/app.log"
  
tracing:
  enabled: true
  mode: "langsmith"
  project: "production-workflows"

Multi-Cloud Integration​

# Advanced configuration with multiple cloud providers
storage_config_path: "agentmap_cloud_storage.yaml"

llm:
  openai:
    api_key: "env:OPENAI_API_KEY"
  anthropic:
    api_key: "env:ANTHROPIC_API_KEY"

routing:
  enabled: true
  routing_matrix:
    anthropic:
      low: "claude-3-haiku-20240307"
      medium: "claude-3-5-sonnet-20241022"
      high: "claude-3-opus-20240229"
    openai:
      low: "gpt-3.5-turbo"
      medium: "gpt-4-turbo"
      high: "gpt-4"

🔧 Configuration Validation​

AgentMap uses Pydantic models for configuration validation, ensuring:

  • ✅ Type Safety: All configuration values are properly typed and validated
  • ✅ Required Fields: Missing required configuration triggers clear error messages
  • ✅ Default Values: Sensible defaults for all optional configuration
  • ✅ Range Validation: Numeric values are validated against acceptable ranges
  • ✅ Enum Validation: String values are validated against allowed options

Validation Example​

# Configuration is automatically validated on startup
from agentmap.services.config import AppConfigService

config_service = AppConfigService("agentmap_config.yaml")
# ✅ Validates all configuration against Pydantic models
# ❌ Raises clear errors for invalid configuration

🚨 Security Best Practices​

Environment Variable Usage​

  • Never put sensitive credentials directly in YAML files
  • Always use env:VARIABLE_NAME syntax for API keys and secrets
  • Use separate environment files for different deployment stages

File Permissions​

# Secure configuration file permissions
chmod 600 agentmap_config.yaml
chmod 600 agentmap_config_storage.yaml
chmod 600 .env

Production Security​

  • Use dedicated service accounts for cloud storage access
  • Implement proper authentication and authorization policies
  • Enable tracing and monitoring for security auditing
  • Regular rotation of API keys and access credentials

📈 Performance Configuration​

Cost Optimization​

  • Enable intelligent routing to use cost-effective models when appropriate
  • Configure cost tier limits to prevent unexpected expenses
  • Use caching to reduce redundant API calls

Resource Management​

  • Configure memory buffers based on workflow complexity
  • Set appropriate timeout and retry policies
  • Enable performance tracing for optimization

Scalability Settings​

  • Configure connection pooling for storage providers
  • Set appropriate concurrency limits
  • Enable request batching where supported

🔄 Configuration Updates​

Hot Reloading​

Some configuration changes can be applied without restart:

  • Environment variable changes
  • Routing policy adjustments
  • Logging level modifications

Restart Required​

Other changes require application restart:

  • Storage provider configuration
  • Authentication settings
  • Core path modifications

📖 Next Steps​

  1. Start with Main Configuration - Understand the complete YAML structure
  2. Configure Storage - Set up your data persistence layer
  3. Set Environment Variables - Secure your credentials
  4. Review Examples - See working configurations for your use case
  5. Check Troubleshooting - Resolve common configuration issues

Ready to configure AgentMap for your specific needs? Choose the section that matches your current requirements and follow the detailed guides.