Skip to content

Configuration

ALMA is configured via a YAML file, typically at .alma/config.yaml.

Basic Configuration

alma:
  project_id: "my-project"
  storage: sqlite           # Storage backend
  embedding_provider: local # Embedding provider
  storage_dir: .alma        # Where to store data
  db_name: alma.db          # Database filename
  embedding_dim: 384        # Embedding dimensions

  agents:
    qa_tester:
      domain: coding
      can_learn:
        - testing_strategies
        - selector_patterns
      cannot_learn:
        - backend_logic
      min_occurrences_for_heuristic: 3

Storage Backends

SQLite (Local Development)

storage: sqlite
storage_dir: .alma
db_name: alma.db

PostgreSQL (Production)

storage: postgres
postgres:
  host: localhost
  port: 5432
  database: alma
  user: alma_user
  password: ${POSTGRES_PASSWORD}  # Environment variable
  vector_index_type: hnsw         # hnsw or ivfflat

Qdrant

storage: qdrant
qdrant:
  url: http://localhost:6333
  api_key: ${QDRANT_API_KEY}  # Optional for cloud
  collection_prefix: alma

Pinecone

storage: pinecone
pinecone:
  api_key: ${PINECONE_API_KEY}
  environment: us-east-1-aws
  index_name: alma-memories

Chroma

storage: chroma
chroma:
  persist_directory: .alma/chroma
  # Or for client-server mode:
  # host: localhost
  # port: 8000

Azure Cosmos DB

storage: azure
azure:
  endpoint: ${COSMOS_ENDPOINT}
  key: ${COSMOS_KEY}
  database_name: alma
  container_name: memories

Embedding Providers

Provider Model Dimensions Config
local all-MiniLM-L6-v2 384 Free, offline
azure text-embedding-3-small 1536 Requires API key
mock Hash-based 384 Testing only 
embedding_provider: local  # or azure, mock
embedding_dim: 384         # Must match provider

Agent Configuration

Basic Agent

agents:
  my-agent:
    domain: general
    can_learn:
      - strategies
      - patterns
    cannot_learn: []
    min_occurrences_for_heuristic: 3

Scoped Learning

agents:
  frontend-tester:
    domain: coding
    can_learn:
      - testing_strategies
      - selector_patterns
      - form_validation
    cannot_learn:
      - backend_logic
      - database_queries
      - api_design

Multi-Agent Sharing

agents:
  senior_dev:
    can_learn: [architecture, best_practices]
    share_with: [junior_dev, qa_agent]  # Others can read my memories

  junior_dev:
    can_learn: [coding_patterns]
    inherit_from: [senior_dev]  # I can read senior's memories

  qa_agent:
    can_learn: [testing_strategies]
    inherit_from: [senior_dev]

Environment Variables

Use ${VAR_NAME} syntax to reference environment variables:

postgres:
  password: ${POSTGRES_PASSWORD}

pinecone:
  api_key: ${PINECONE_API_KEY}

Complete Example

alma:
  project_id: "production-app"
  storage: postgres
  embedding_provider: azure
  embedding_dim: 1536

  postgres:
    host: ${DB_HOST}
    port: 5432
    database: alma_prod
    user: alma
    password: ${DB_PASSWORD}
    vector_index_type: hnsw

  azure_openai:
    endpoint: ${AZURE_OPENAI_ENDPOINT}
    api_key: ${AZURE_OPENAI_KEY}
    deployment_name: text-embedding-3-small

  agents:
    qa_tester:
      domain: coding
      can_learn:
        - testing_strategies
        - selector_patterns
        - accessibility_checks
      cannot_learn:
        - backend_logic
      min_occurrences_for_heuristic: 3
      share_with: [qa_lead]

    backend_dev:
      domain: coding
      can_learn:
        - api_patterns
        - database_queries
        - performance_optimization
      cannot_learn:
        - frontend_selectors
      inherit_from: [senior_architect]

    senior_architect:
      domain: coding
      can_learn:
        - architecture_decisions
        - best_practices
        - security_patterns
      share_with: [qa_tester, backend_dev]

Next Steps