The config.yaml file is a configuration file that stores all the environment variables for the application. This file should be placed at the root of the /app directory. It can have multiple environments such as dev, prod, etc. Each environment should be placed under a key with the name of the environment, and the key must be at the top level of the file. The environment that is used is determined by the global_env key-value pair (e.g. global_env: dev). It’s important to note that the variables in this file are converted into environment variables. Essentially, every key-value pair at the leaf level is extracted. The key becomes the key of the environment variable in all caps, and the value remains the same.

Example: The microsoft_file_picker_client_id variable under prod.data_connects.onedrive would be converted to the env variable: MICROSOFT_CLIENT_FILE_PICKER_CLIENT_ID=test_id_here.

You can also pass these variables in as an .env file or simply as regular environment variables. 
global_env: dev
dev:
  general:
    env: '' # local, dev, or prod
    data_source_encryption_key: ''
    hash_bearer_token: '' # true or false
    api_hostname: ''
    auth_access_token_secret: ''
    auth_refresh_token_secret: ''
    disable_api: '' # true or false
    disable_background_processes: '' # true or false
    disable_resync: '' # true or false
    disable_textract: '' # true or false
    queue_backend: '' # Queue backend (PUBSUB , SQS, ElasticMQ, or NATS)
    send_emails: '' # Control whether to send emails or not (true or false). Defaults to true.
    use_custom_queue: '' # If using a custom queue, provide the path to the file it's implemented in.
    vertex_service_account_key: '' # For local dev
    carbon_webhook_signing_key: '' # for testing carbon webhooks
    disable_rate_limits: '' # true or false
    disable_protected_routes: '' # default true, can be set to false to enable certain routes in a safe env
    disable_autosync_service: '' # default false, can be set to true to disable autosyncing files from 3rd party connectors
  request_timeouts:
    default_request_timeout: 7 # used for all requests made to external APIs and URLs by default
    web_scrape_request_timeout: 60 # used for requests made to apify during web scrape
    <data_source_name>_request_timeout: 7 # used to customise request timeout for specific data sources. Replace data_source_name with actual name, for example notion_request_timeout and google_drive_request_timeout
  storage:
    storage_backend: '' # Storage backend (S3, S3-compatible, GCS, Azure, or local)
    storage_endpoint_url: '' # only relevant for S3-compatible storage
    storage_access_key_id: '' # only relevant for S3-compatible storage
    storage_secret_access_key: '' # only relevant for S3-compatible storage
    storage_tls: '' # only relevant for S3-compatible storage (true or false)
    storage_sse: '' # only relevant for S3-compatible storage (true or false)
    azure_storage_account_url: '' # only relevant for Azure storage
    gcs_service_account_email: '' # required if using GCP with a service account
  logging:
    better_exceptions: 1
    log_json_format: '' # true or false
    log_level: '' # INFO or DEBUG
    no_logging: '' # true or false
  db:
    db_host: ''
    db_host_reader: ''
    db_name: ''
    db_name_reader: ''
    db_password: ''
    db_password_reader: ''
    db_port: ''
    db_port_reader: ''
    db_schema: ''
    db_schema_reader: ''
    db_user: ''
    db_user_reader: ''
    instance_db_pool_overflow: ''
    instance_db_pool_size: ''
  redis:
    # only redis_db and redis_host are requried
    redis_db: ''
    redis_host: ''
    redis_password: ''
    redis_port: ''
  vector_db:
    qdrant_api_key: ''
    qdrant_bm25_collection_name: ''
    qdrant_cohere_multilingual_v3_collection_name: ''
    qdrant_collection_name: ''
    qdrant_openai_ada_large_1024_collection_name: ''
    qdrant_openai_ada_large_256_collection_name: ''
    qdrant_openai_ada_large_3072_collection_name: ''
    qdrant_openai_ada_small_1536_collection_name: ''
    qdrant_openai_ada_small_512_collection_name: ''
    qdrant_solar_1_mini_collection_name: ''
    qdrant_url: ''
    qdrant_vertex_multimodal_collection_name: ''
  s3_buckets:
    rss_feeds_bucket: ''
    s3_cold_storage_bucket: ''
    s3_config_bucket: ''
    s3_customer_files_bucket: ''
    s3_dictionary_and_tfidf_bucket: ''
  gcs_buckets:
    gcs_customer_videos_bucket: ''
  queues:
    # only required if using sqs
    sqs_all_uploaded_files_queued_webhook_queue: ''
    sqs_bm25_queue: ''
    sqs_cold_storage_webhook_queue: ''
    sqs_cold_to_hot_storage_queue: ''
    sqs_customer_files_queue: ''
    sqs_data_source_actions_webhook_queue: ''
    sqs_data_source_ready_webhook_queue: ''
    sqs_delete_files_queue: ''
    sqs_delete_files_v2_queue: ''
    sqs_delete_users_queue: ''
    sqs_external_api_customer_files_queue: ''
    sqs_external_api_dropbox_queue: ''
    sqs_external_api_emails_queue: ''
    sqs_external_api_help_center_queue: ''
    sqs_external_api_onedrive_queue: ''
    sqs_external_source_items_queue: ''
    sqs_fetch_user_webpages_queue: ''
    sqs_file_deleted_webhook_queue: ''
    sqs_file_error_webhook_queue: ''
    sqs_file_sync_limit_reached_webhook_queue: ''
    sqs_file_syncing_webhook_queue: ''
    sqs_files_created_webhook_queue: ''
    sqs_google_drive_queue: ''
    sqs_notion_rate_limit_queue: ''
    sqs_organization_user_deleted_webhook_queue: ''
    sqs_rate_limit_error_webhook_queue: ''
    sqs_raw_text_uploads_queue: ''
    sqs_video_uploads_queue: ''
    sqs_webhook_queue: ''
    sqs_webpage_webhooks_queue: ''
    sqs_webscrape_batch_queue: ''
    sqs_webscrape_queue: ''
    sqs_zotero_queue: ''
    sqs_all_files_processed_webhook_queue: ''
    sqs_external_api_slack_queue: ''
    # Dedicated autosync queues can optionally be created for all third party connectors. 
    # The naming convention for env variables will be appending _autosync to the original queue.
    # For example autosync queue for sqs_external_api_dropbox_queue will be sqs_external_api_dropbox_queue_autosync
    sqs_external_api_customer_files_queue_autosync: ''
    sqs_external_api_dropbox_queue_autosync: ''
    sqs_external_api_emails_queue_autosync: ''
    sqs_external_api_help_center_queue_autosync: ''
    sqs_external_api_onedrive_queue_autosync: ''
    sqs_google_drive_queue_autosync: ''
    sqs_notion_rate_limit_queue_autosync: ''
    sqs_external_api_slack_queue_autosync: ''
  pubsub:
    # only required if using pubsub instead of sqs
    pubsub_project_id: ''
    pubsub_topic: ''
  nats:
    # only required if using nats instead of sqs
    nats_connection_string: ''
    use_jetstream: '' # true or false
  embedding_models:
    cohere:
      cohere_api_key: ''
    openai:
      azure_openai_ada_large_deployment_id: ''
      azure_openai_ada_small_deployment_id: ''
      azure_openai_api_base: ''
      azure_openai_api_key: ''
      azure_openai_deployment_id: ''
      azure_openai_v3_api_base: ''
      azure_openai_v3_api_key: ''
      default_ada_large_1024_model: ''
      default_ada_large_256_model: ''
      default_ada_large_3072_model: ''
      default_ada_small_1536_model: ''
      default_ada_small_512_model: ''
      default_query_text_embedding_model: ''
      openai_accepted_latency_seconds: ''
      openai_api_key: ''
    jina:
      jina_api_key: ''
    solar:
      solar_api_key: ''
  data_connectors:
    protect_default_oauth_access_tokens: ''
    atlassian:
      atlassian_client_id: ''
      atlassian_client_secret: ''
    box:
      box_client_id: ''
      box_client_secret: ''
    dropbox:
      alt_2_dropbox_app_key: ''
      alt_2_dropbox_app_secret: ''
      alt_dropbox_app_key: ''
      alt_dropbox_app_secret: ''
      dropbox_app_key: ''
      dropbox_app_secret: ''
    gmail:
      alt_3_gmail_client_id: ''
      alt_3_gmail_client_secret: ''
      alt_2_gmail_client_id: ''
      alt_2_gmail_client_secret: ''
      alt_gmail_client_id: ''
      alt_gmail_client_secret: ''
      gmail_client_id: ''
      gmail_client_secret: ''
    google:
      google_alt_2_api_key: ''
      google_alt_2_client_id: ''
      google_alt_2_client_secret: ''
      google_alt_3_api_key: ''
      google_alt_3_client_id: ''
      google_alt_3_client_secret: ''
      google_alt_4_api_key: ''
      google_alt_4_client_id: ''
      google_alt_4_client_secret: ''
      google_alt_api_key: ''
      google_alt_client_id: ''
      google_alt_client_secret: ''
      google_api_key: ''
      google_client_id: ''
      google_client_secret: ''
      google_redirect_uri: ''
    google_cloud_storage:
      google_cloud_storage_api_key: ''
      google_cloud_storage_client_id: ''
      google_cloud_storage_client_secret: ''
    intercom:
      intercom_client_id: ''
      intercom_client_secret: ''
    notion:
      notion_client_id: ''
      notion_client_secret: ''
      notion_redirect_uri: ''
    one_drive:
      microsoft_file_picker_client_id: ''
      onedrive_client_id: ''
      onedrive_client_secret: ''
    salesforce:
      salesforce_client_id: ''
      salesforce_client_secret: ''
      salesforce_verifier: ''
    slack:
      slack_client_id: ''
      slack_client_secret: ''
    zendesk:
      zendesk_app_subdomain: ''
      zendesk_client_id: ''
      zendesk_client_secret: ''
    zotero:
      zotero_client_key: ''
      zotero_client_secret: ''
    gong:
      gong_client_id: ''
      gong_client_secret: ''
  data_processors:
    apify:
      apify_actor_id: ''
      apify_api_key: ''
      apify_api_key_for_resync: ''
      default_apify_account_max_ram: ''
      apify_slack_alerts_token: ''
      apify_slack_alerts_channel: ''
    sauce_labs:
      sauce_labs_access_key: ''
      sauce_labs_region: ''
      sauce_labs_username: ''
    assemblyai_api_key: ''
    deepgram_api_key: ''
    metaphor_api_key: ''
    youtube_transcript_proxy: ''
  loops:
    loops_api_key: ''
  paigo:
    paigo_chars_stored_dimension_id: ''
    paigo_chars_synced_dimension_id: ''
    paigo_client_id: ''
    paigo_client_secret: ''
    paigo_embedding_dimension_id: ''
  stripe:
    stripe_api_key: ''
    stripe_bytes_uploaded_event_name: ''
    stripe_characters_synced_event_name: ''
    stripe_pages_ocr_scanned_event_name: ''
    stripe_urls_scraped_event_name: ''
  supabase:
    supabase_host: ''
    supabase_name: ''
    supabase_password: ''
    supabase_port: ''
    supabase_service_key: ''
    supabase_url: ''
    supabase_user: ''