Configure Cogitator with YAML files, environment variables, and provider settings.

Configuration File

Create a cogitator.yml in your project root:

llm:
  defaultProvider: ollama
  providers:
    ollama:
      type: ollama
      host: http://localhost:11434
      model: llama3.2
    openai:
      type: openai
      apiKey: ${OPENAI_API_KEY}
      model: gpt-4o
    anthropic:
      type: anthropic
      apiKey: ${ANTHROPIC_API_KEY}
      model: claude-sonnet-4-5-20250929

memory:
  adapter: redis
  redis:
    url: redis://localhost:6379
  embedding:
    provider: ollama
    model: nomic-embed-text

logging:
  level: info
  format: pretty

Load it with @cogitator-ai/config:

import { Cogitator } from '@cogitator-ai/core';
import { loadConfig } from '@cogitator-ai/config';

const config = await loadConfig();
const cogitator = new Cogitator(config);

loadConfig() automatically:

Reads cogitator.yml from the current working directory
Expands ${ENV_VAR} references from environment variables
Validates the config against the expected schema
Merges with sensible defaults

Environment Variables

Store secrets in .env:

# LLM Providers
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
GOOGLE_API_KEY=AIza...
AZURE_OPENAI_API_KEY=...
AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com

# Infrastructure
REDIS_URL=redis://localhost:6379
DATABASE_URL=postgresql://user:pass@localhost:5432/cogitator

# Observability
LANGFUSE_PUBLIC_KEY=pk-...
LANGFUSE_SECRET_KEY=sk-...

Programmatic Configuration

You can skip the YAML file and configure directly:

const cogitator = new Cogitator({
  llm: {
    defaultProvider: 'openai',
    providers: {
      openai: {
        type: 'openai',
        apiKey: process.env.OPENAI_API_KEY!,
        model: 'gpt-4o',
      },
    },
  },
  memory: {
    adapter: 'postgres',
    connectionString: process.env.DATABASE_URL!,
    embedding: {
      provider: 'openai',
      model: 'text-embedding-3-small',
    },
  },
});

Provider Configuration

Ollama (Local)

providers:
  ollama:
    type: ollama
    host: http://localhost:11434
    model: llama3.2

OpenAI

providers:
  openai:
    type: openai
    apiKey: ${OPENAI_API_KEY}
    model: gpt-4o
    organization: org-... # optional

Anthropic

providers:
  anthropic:
    type: anthropic
    apiKey: ${ANTHROPIC_API_KEY}
    model: claude-sonnet-4-5-20250929

Google Gemini

providers:
  google:
    type: google
    apiKey: ${GOOGLE_API_KEY}
    model: gemini-2.0-flash

Azure OpenAI

providers:
  azure:
    type: azure
    apiKey: ${AZURE_OPENAI_API_KEY}
    endpoint: ${AZURE_OPENAI_ENDPOINT}
    deployment: gpt-4o
    apiVersion: '2024-02-15-preview'

AWS Bedrock

providers:
  bedrock:
    type: bedrock
    region: us-east-1
    model: anthropic.claude-3-sonnet-20240229-v1:0
    # Uses AWS SDK default credential chain

Config Merging

Configs are merged with this priority (highest wins):

Programmatic options passed to new Cogitator({...})
Environment variables (${ENV_VAR} expansions in YAML)
cogitator.yml file values
Built-in defaults

Configuration

On this page