Configuration Reference - Omni Documentation

This page provides a comprehensive reference for all environment variables and configuration options available in Omni.

All environment variables can be set in your .env file for Docker Compose deployments or in terraform.tfvars for AWS deployments.

Quick Configuration Examples

Minimal Configuration (Development)

# Database
DATABASE_PASSWORD=your-secure-password

# Security
SESSION_SECRET=generate-32-char-secret
ENCRYPTION_KEY=generate-32-char-key
ENCRYPTION_SALT=generate-16-char-salt

# Application
APP_URL=http://localhost:3000

# LLM Provider
LLM_PROVIDER=anthropic
ANTHROPIC_API_KEY=sk-ant-your-key

# Embeddings
EMBEDDING_PROVIDER=jina
JINA_API_KEY=your-jina-key

Production Configuration

# Database
DATABASE_HOST=postgres
DATABASE_PORT=5432
DATABASE_NAME=omni
DATABASE_USER=omni
DATABASE_PASSWORD=your-secure-password
DATABASE_SSL=true
DB_MAX_CONNECTIONS=20

# Redis
REDIS_URL=redis://redis:6379

# Security
SESSION_SECRET=your-32-char-secret
SESSION_DURATION_DAYS=30
ENCRYPTION_KEY=your-32-char-key
ENCRYPTION_SALT=your-16-char-salt

# Application
APP_URL=https://omni.yourcompany.com
NODE_ENV=production
LOG_LEVEL=info

# LLM Provider (Self-hosted)
LLM_PROVIDER=vllm
VLLM_URL=http://vllm:8000
VLLM_MODEL=meta-llama/Llama-3.1-8B-Instruct

# Embeddings
EMBEDDING_PROVIDER=jina
JINA_API_KEY=your-jina-key

# Storage
STORAGE_BACKEND=s3
S3_BUCKET=omni-content
S3_REGION=us-east-1

# Email
EMAIL_PROVIDER=resend
EMAIL_FROM=Omni <[email protected]>
RESEND_API_KEY=re_your-key

# HTTPS
ACME_EMAIL=[email protected]
CLIO_DOMAIN=omni.yourcompany.com

Database Configuration

Variable	Required	Default	Description
`DATABASE_HOST`	Yes	`postgres`	Database hostname or IP address
`DATABASE_PORT`	Yes	`5432`	Database port
`DATABASE_USERNAME`	Yes	`omni`	Database username
`DATABASE_PASSWORD`	Yes	-	Database password (use strong password)
`DATABASE_NAME`	Yes	`omni`	Database name
`DATABASE_SSL`	No	`false`	Enable SSL for database connection
`DB_MAX_CONNECTIONS`	No	`10`	Connection pool size per service
`DB_ACQUIRE_TIMEOUT_SECONDS`	No	`3`	Connection acquisition timeout

Generate a secure database password:

openssl rand -base64 32

Redis Configuration

Variable	Required	Default	Description
`REDIS_URL`	Yes	`redis://redis:6379`	Redis connection URL (format: `redis://host:port`)

For Redis with password:

REDIS_URL=redis://:password@redis:6379

Application Configuration

Variable	Required	Default	Description
`APP_URL`	Yes	`http://localhost:3000`	Public-facing application URL (include protocol)
`OMNI_DOMAIN`	No	`localhost`	Domain name for the application
`SESSION_SECRET`	Yes	-	Secret key for session encryption (32+ characters)
`SESSION_COOKIE_NAME`	No	`auth-session`	Name of the session cookie
`SESSION_DURATION_DAYS`	No	`7`	Session expiry in days
`NODE_ENV`	No	`development`	Node environment: `development` or `production`

Never use the same SESSION_SECRET across different environments. Generate unique secrets for dev, staging, and production.

Security & Encryption

Variable	Required	Default	Description
`ENCRYPTION_KEY`	Yes	-	Encryption key for OAuth tokens (32+ characters)
`ENCRYPTION_SALT`	Yes	-	Salt for key derivation (16+ characters)

Generate secure encryption keys:

# For ENCRYPTION_KEY (32 characters)
openssl rand -hex 16

# For ENCRYPTION_SALT (16 characters)
openssl rand -hex 8

# For SESSION_SECRET (32 characters)
openssl rand -hex 16

Service Ports

Variable	Required	Default	Description
`WEB_PORT`	No	`3000`	SvelteKit web application port
`SEARCHER_PORT`	No	`3001`	Search service port
`INDEXER_PORT`	No	`3002`	Indexer service port
`AI_SERVICE_PORT`	No	`3003`	AI service port
`GOOGLE_CONNECTOR_PORT`	No	`4001`	Google connector port
`SLACK_CONNECTOR_PORT`	No	`4002`	Slack connector port
`ATLASSIAN_CONNECTOR_PORT`	No	`4003`	Atlassian connector port
`VLLM_PORT`	No	`8000`	vLLM inference server port

In Docker Compose, services communicate via service names (e.g., http://searcher:3001). Ports only need to be exposed to the host for debugging.

Inter-Service URLs

Variable	Required	Default	Description
`SEARCHER_URL`	Yes	`http://searcher:3001`	Search service URL
`INDEXER_URL`	Yes	`http://indexer:3002`	Indexer service URL
`AI_SERVICE_URL`	Yes	`http://ai:3003`	AI service URL
`GOOGLE_CONNECTOR_URL`	Yes	`http://google-connector:4001`	Google connector URL

Feature Flags

Variable	Required	Default	Description
`AI_ANSWER_ENABLED`	No	`true`	Enable AI-generated answers in search results
`AI_FIRST_SEARCH_ENABLED`	No	`true`	Display AI answers before traditional search results

LLM Provider Configuration

Variable	Required	Default	Description
`LLM_PROVIDER`	Yes	`vllm`	LLM provider: `vllm`, `anthropic`, or `bedrock`

vLLM (Self-Hosted)

For self-hosted LLM inference:

Variable	Required	Default	Description
`VLLM_URL`	Yes	`http://vllm:8000`	vLLM service URL
`VLLM_MODEL`	No	`microsoft/Phi-3-mini-4k-instruct`	HuggingFace model identifier
`VLLM_DTYPE`	No	`auto`	Data type: `auto`, `float16`, `bfloat16`
`VLLM_API_KEY`	No	`token-abc123`	API key for vLLM authentication
`VLLM_MAX_MODEL_LEN`	No	`2048`	Maximum context length in tokens

Popular model options:

microsoft/Phi-3-mini-4k-instruct (4 GB VRAM)
mistralai/Mistral-7B-Instruct-v0.2 (14 GB VRAM)
meta-llama/Llama-3.1-8B-Instruct (16 GB VRAM)

Anthropic Claude

For Anthropic’s Claude API:

Variable	Required	Default	Description
`ANTHROPIC_API_KEY`	Yes	-	Anthropic API key from https://console.anthropic.com
`ANTHROPIC_MODEL`	No	`claude-sonnet-4-20250514`	Claude model version
`ANTHROPIC_MAX_TOKENS`	No	`4096`	Maximum tokens in response

Available models:

claude-sonnet-4-20250514 (Recommended)
claude-opus-4-20250514 (Highest quality)
claude-haiku-4-20250901 (Fastest, lowest cost)

AWS Bedrock

For AWS Bedrock:

Variable	Required	Default	Description
`BEDROCK_MODEL_ID`	No	`us.anthropic.claude-sonnet-4-20250514-v1:0`	Bedrock model ID
`TITLE_GENERATION_MODEL_ID`	No	`us.anthropic.claude-haiku-4-5-20251001-v1:0`	Model for title generation
`AWS_REGION`	Yes	-	AWS region for Bedrock (e.g., `us-east-1`)
`AWS_ACCESS_KEY_ID`	Conditional	-	AWS access key (if not using IAM role)
`AWS_SECRET_ACCESS_KEY`	Conditional	-	AWS secret key (if not using IAM role)

AI Service Configuration

Variable	Required	Default	Description
`MODEL_PATH`	No	`/models`	Directory for model storage
`EMBEDDING_MODEL`	No	`intfloat/e5-large-v2`	HuggingFace embedding model (if using local)
`EMBEDDING_DIMENSIONS`	No	`1024`	Dimension of embedding vectors
`EMBEDDING_PROVIDER`	Yes	`jina`	Embedding provider: `jina`, `local`, or `bedrock`
`JINA_API_KEY`	Conditional	-	Jina AI API key (required if `EMBEDDING_PROVIDER=jina`)
`AI_WORKERS`	No	`1`	Number of uvicorn worker processes
`CUDA_VISIBLE_DEVICES`	No	`0`	GPU device ID for local models (comma-separated for multiple)

For Jina embeddings, get a free API key at https://jina.ai. The free tier includes 1M tokens/month.

Conversation Compaction

For long conversations that exceed LLM context limits, Omni automatically compacts older messages into a summary while preserving recent messages.

Variable	Required	Default	Description
`ENABLE_CONVERSATION_COMPACTION`	No	`true`	Enable automatic conversation compaction
`MAX_CONVERSATION_INPUT_TOKENS`	No	`150000`	Maximum token limit (compaction triggers at 80%)
`COMPACTION_RECENT_MESSAGES_COUNT`	No	`20`	Number of recent messages to keep intact
`COMPACTION_SUMMARY_MAX_TOKENS`	No	`2000`	Maximum tokens for the generated summary
`COMPACTION_CACHE_TTL_SECONDS`	No	`86400`	Cache TTL for summaries (24 hours)

Compaction triggers when conversation tokens reach 80% of MAX_CONVERSATION_INPUT_TOKENS (default: 120,000 tokens). The compactor preserves tool_use/tool_result pairs and caches summaries in Redis.

Batch Embedding Inference (AWS Bedrock)

For large-scale embedding generation using AWS Bedrock batch inference:

Variable	Required	Default	Description
`ENABLE_EMBEDDING_BATCH_INFERENCE`	No	`false`	Enable batch processing for embeddings
`EMBEDDING_BATCH_S3_BUCKET`	Conditional	-	S3 bucket for batch files (required if batch enabled)
`EMBEDDING_BATCH_BEDROCK_ROLE_ARN`	Conditional	-	IAM role ARN for Bedrock (required if batch enabled)
`EMBEDDING_BATCH_MIN_DOCUMENTS`	No	`100`	Minimum documents to trigger batch job
`EMBEDDING_BATCH_MAX_DOCUMENTS`	No	`50000`	Maximum documents per batch
`EMBEDDING_BATCH_ACCUMULATION_TIMEOUT_SECONDS`	No	`300`	Wait time before starting batch (5 min)
`EMBEDDING_BATCH_ACCUMULATION_POLL_INTERVAL`	No	`10`	Interval to check queue (10 sec)
`EMBEDDING_BATCH_MONITOR_POLL_INTERVAL`	No	`300`	Interval to check batch status (5 min)

Google Workspace Connector

Variable	Required	Default	Description
`GOOGLE_CLIENT_ID`	Yes	-	Google OAuth 2.0 client ID
`GOOGLE_CLIENT_SECRET`	Yes	-	Google OAuth 2.0 client secret
`GOOGLE_REDIRECT_URI`	Yes	-	OAuth callback URL (e.g., `https://omni.company.com/api/oauth/google/callback`)
`GOOGLE_SYNC_INTERVAL_SECONDS`	No	`86400`	Interval between full syncs (24 hours)
`GOOGLE_MAX_AGE_DAYS`	No	`730`	Maximum age of files to index (2 years)
`GOOGLE_WEBHOOK_URL`	No	-	HTTPS URL for real-time updates (requires valid SSL certificate)
`WEBHOOK_RENEWAL_CHECK_INTERVAL_SECONDS`	No	`3600`	Interval to check and renew webhooks (1 hour)

To set up Google OAuth credentials, visit the Google Cloud Console. See Google Connector Setup for detailed instructions.

Slack Connector

Variable	Required	Default	Description
`SLACK_CLIENT_ID`	Yes	-	Slack OAuth client ID
`SLACK_CLIENT_SECRET`	Yes	-	Slack OAuth client secret
`SLACK_BOT_TOKEN`	Yes	-	Slack bot token (starts with `xoxb-`)

Create a Slack app at https://api.slack.com/apps to get credentials. See Slack Connector Setup for detailed instructions.

Atlassian Connector

Variable	Required	Default	Description
`ATLASSIAN_CLIENT_ID`	Yes	-	Atlassian OAuth client ID
`ATLASSIAN_CLIENT_SECRET`	Yes	-	Atlassian OAuth client secret
`ATLASSIAN_BASE_URL`	Yes	-	Atlassian instance URL (e.g., `https://yourcompany.atlassian.net`)
`ATLASSIAN_USER_EMAIL`	Yes	-	User email for API authentication
`ATLASSIAN_API_TOKEN`	Yes	-	Atlassian API token

Generate an API token at https://id.atlassian.com/manage-profile/security/api-tokens. See Atlassian Connector Setup for detailed instructions.

HubSpot Connector

Variable	Required	Default	Description
`HUBSPOT_CONNECTOR_PORT`	No	`4004`	HubSpot connector service port

The HubSpot connector requires credentials to be configured through the Omni UI:

Private App Token: A HubSpot private app token (starts with pat-na1-)
Portal ID: Your HubSpot portal ID (optional)

Required OAuth scopes for Private Apps:

crm.objects.contacts.read
crm.objects.companies.read
crm.objects.deals.read
tickets
e-commerce (for engagements)

See HubSpot Connector Setup for detailed instructions.

Filesystem Connector

Variable	Required	Default	Description
`FILESYSTEM_CONNECTOR_PORT`	No	`4006`	Filesystem connector service port

The Filesystem connector is configured through source settings:

base_path: Root directory path to scan (required)
scan_interval_seconds: Full scan interval (default: 300)
file_extensions: Whitelist of extensions to include (optional)
exclude_patterns: Path patterns to exclude (optional)
max_file_size_bytes: Maximum file size to process (default: 10MB for content)

See Filesystem Connector Setup for detailed instructions including Docker volume mounting.

Connector Manager

The connector-manager service orchestrates all connector operations.

Variable	Required	Default	Description
`CONNECTOR_MANAGER_PORT`	No	`8090`	Connector manager service port
`MAX_CONCURRENT_SYNCS`	No	`10`	Maximum concurrent syncs across all sources
`MAX_CONCURRENT_SYNCS_PER_TYPE`	No	`3`	Maximum concurrent syncs per connector type
`SCHEDULER_INTERVAL_SECONDS`	No	`30`	How often scheduler checks for due syncs
`STALE_SYNC_TIMEOUT_MINUTES`	No	`10`	Timeout to mark a sync as stale/failed

Connector URLs (used by connector-manager to communicate with connectors):

Variable	Required	Default	Description
`CONNECTOR_GOOGLE_URL`	Yes	-	Google connector URL
`CONNECTOR_SLACK_URL`	Yes	-	Slack connector URL
`CONNECTOR_ATLASSIAN_URL`	Yes	-	Atlassian connector URL
`CONNECTOR_WEB_URL`	Yes	-	Web connector URL

Storage Configuration

Variable	Required	Default	Description
`STORAGE_BACKEND`	Yes	`s3`	Storage backend: `s3` or `local`
`S3_BUCKET`	Conditional	-	S3 bucket name (required if `STORAGE_BACKEND=s3`)
`S3_REGION`	Conditional	-	S3 region (required if `STORAGE_BACKEND=s3`)
`AWS_ACCESS_KEY_ID`	Conditional	-	AWS access key (if not using IAM role)
`AWS_SECRET_ACCESS_KEY`	Conditional	-	AWS secret key (if not using IAM role)
`AWS_SESSION_TOKEN`	No	-	AWS session token (for temporary credentials)

Local storage example:

STORAGE_BACKEND=local
# Files stored in /data/storage directory

S3 storage example:

STORAGE_BACKEND=s3
S3_BUCKET=omni-content-prod
S3_REGION=us-east-1
# Use IAM role in AWS, or provide keys for other environments

Email Configuration

Variable	Required	Default	Description
`EMAIL_PROVIDER`	Yes	`resend`	Email provider: `resend` or `smtp`
`EMAIL_FROM`	Yes	`Omni <[email protected]>`	From email address

Resend Provider

Variable	Required	Default	Description
`RESEND_API_KEY`	Yes	-	Resend API key from https://resend.com

SMTP Provider

Variable	Required	Default	Description
`EMAIL_HOST`	Yes	-	SMTP server hostname
`EMAIL_PORT`	Yes	`587`	SMTP server port (587 for TLS, 465 for SSL)
`EMAIL_USER`	Yes	-	SMTP username
`EMAIL_PASSWORD`	Yes	-	SMTP password
`EMAIL_SECURE`	No	`true`	Use TLS/SSL

Search Configuration

Variable	Required	Default	Description
`TYPO_TOLERANCE_ENABLED`	No	`true`	Enable typo tolerance in search queries
`TYPO_TOLERANCE_MAX_DISTANCE`	No	`2`	Maximum Levenshtein distance for typo correction
`TYPO_TOLERANCE_MIN_WORD_LENGTH`	No	`4`	Minimum word length to apply typo correction

Logging & Monitoring

Variable	Required	Default	Description
`RUST_LOG`	No	`info`	Rust services log level: `trace`, `debug`, `info`, `warn`, `error`
`LOG_LEVEL`	No	`debug`	Python services log level: `DEBUG`, `INFO`, `WARNING`, `ERROR`

Log level recommendations:

Development: RUST_LOG=debug, LOG_LEVEL=DEBUG
Production: RUST_LOG=info, LOG_LEVEL=INFO
Troubleshooting: RUST_LOG=trace, LOG_LEVEL=DEBUG

Telemetry (OpenTelemetry)

Variable	Required	Default	Description
`OTEL_EXPORTER_OTLP_ENDPOINT`	No	-	OTLP collector endpoint (empty = telemetry disabled)
`OTEL_DEPLOYMENT_ID`	No	-	Deployment identifier for tracing
`OTEL_DEPLOYMENT_ENVIRONMENT`	No	`development`	Environment: `development`, `staging`, `production`
`SERVICE_VERSION`	No	`0.1.0`	Service version for tracing

Example with Honeycomb:

OTEL_EXPORTER_OTLP_ENDPOINT=https://api.honeycomb.io
OTEL_EXPORTER_OTLP_HEADERS=x-honeycomb-team=your-api-key
OTEL_DEPLOYMENT_ID=omni-prod-us-east-1
OTEL_DEPLOYMENT_ENVIRONMENT=production

HTTPS & SSL (Production)

Variable	Required	Default	Description
`ACME_EMAIL`	Conditional	-	Email for Let’s Encrypt notifications (required for automatic HTTPS)
`CLIO_DOMAIN`	Conditional	-	Domain name for Caddy (required for automatic HTTPS)

Automatic HTTPS setup:

ACME_EMAIL=[email protected]
CLIO_DOMAIN=omni.yourcompany.com

Caddy will automatically:

Obtain SSL certificate from Let’s Encrypt
Configure HTTPS
Redirect HTTP to HTTPS
Auto-renew certificates

Environment-Specific Examples

Development (.env)

# Minimal configuration for local development
DATABASE_PASSWORD=dev-password
SESSION_SECRET=dev-session-secret-32-characters
ENCRYPTION_KEY=dev-encryption-key-32-chars
ENCRYPTION_SALT=dev-salt-16char
APP_URL=http://localhost:3000

LLM_PROVIDER=anthropic
ANTHROPIC_API_KEY=sk-ant-your-dev-key

EMBEDDING_PROVIDER=jina
JINA_API_KEY=jina_your-dev-key

RUST_LOG=debug
LOG_LEVEL=DEBUG

Staging (.env)

# Staging environment with production-like settings
DATABASE_HOST=staging-db.internal
DATABASE_PASSWORD=staging-secure-password
DATABASE_SSL=true

SESSION_SECRET=staging-session-secret-unique-32-chars
ENCRYPTION_KEY=staging-encryption-key-unique-32
ENCRYPTION_SALT=staging-salt-16

APP_URL=https://omni-staging.yourcompany.com

LLM_PROVIDER=anthropic
ANTHROPIC_API_KEY=sk-ant-your-staging-key

EMBEDDING_PROVIDER=jina
JINA_API_KEY=jina_your-staging-key

STORAGE_BACKEND=s3
S3_BUCKET=omni-content-staging
S3_REGION=us-east-1

EMAIL_PROVIDER=resend
RESEND_API_KEY=re_staging_key

RUST_LOG=info
LOG_LEVEL=INFO

OTEL_EXPORTER_OTLP_ENDPOINT=https://api.honeycomb.io
OTEL_DEPLOYMENT_ENVIRONMENT=staging

Production (.env)

# Production configuration with all features
DATABASE_HOST=prod-db.internal
DATABASE_PORT=5432
DATABASE_NAME=omni
DATABASE_USER=omni
DATABASE_PASSWORD=prod-very-secure-password-here
DATABASE_SSL=true
DB_MAX_CONNECTIONS=20

REDIS_URL=redis://:redis-password@redis:6379

SESSION_SECRET=prod-session-secret-unique-32-characters
SESSION_DURATION_DAYS=30
ENCRYPTION_KEY=prod-encryption-key-unique-32-chars
ENCRYPTION_SALT=prod-salt-unique

APP_URL=https://omni.yourcompany.com
NODE_ENV=production

LLM_PROVIDER=vllm
VLLM_URL=http://vllm:8000
VLLM_MODEL=meta-llama/Llama-3.1-8B-Instruct

EMBEDDING_PROVIDER=jina
JINA_API_KEY=jina_prod_key

STORAGE_BACKEND=s3
S3_BUCKET=omni-content-prod
S3_REGION=us-east-1

EMAIL_PROVIDER=resend
EMAIL_FROM=Omni <[email protected]>
RESEND_API_KEY=re_prod_key

GOOGLE_CLIENT_ID=your-prod-client-id.apps.googleusercontent.com
GOOGLE_CLIENT_SECRET=GOCSPX-your-prod-secret
GOOGLE_REDIRECT_URI=https://omni.yourcompany.com/api/oauth/google/callback
GOOGLE_WEBHOOK_URL=https://omni.yourcompany.com/api/webhooks/google

SLACK_CLIENT_ID=your-slack-client-id
SLACK_CLIENT_SECRET=your-slack-client-secret
SLACK_BOT_TOKEN=xoxb-your-bot-token

ACME_EMAIL=[email protected]
CLIO_DOMAIN=omni.yourcompany.com

RUST_LOG=info
LOG_LEVEL=INFO

OTEL_EXPORTER_OTLP_ENDPOINT=https://api.honeycomb.io
OTEL_DEPLOYMENT_ID=omni-prod-us-east-1
OTEL_DEPLOYMENT_ENVIRONMENT=production

Security Best Practices

Generate Strong Secrets

Use cryptographically secure random values for all secrets:

# Generate all secrets at once
cat > .env.secrets <<EOF
SESSION_SECRET=$(openssl rand -hex 16)
ENCRYPTION_KEY=$(openssl rand -hex 16)
ENCRYPTION_SALT=$(openssl rand -hex 8)
DATABASE_PASSWORD=$(openssl rand -base64 32)
EOF

# Then copy values to .env

Protect Environment Files

# Never commit .env to version control
echo ".env" >> .gitignore
echo ".env.*" >> .gitignore

# Set restrictive permissions
chmod 600 .env

# For production, use secret management systems
# - AWS Secrets Manager (Terraform deployments)
# - Docker Secrets (Swarm deployments)
# - Kubernetes Secrets (K8s deployments)

Rotate Secrets Regularly

Best practices for secret rotation:

API Keys: Rotate every 90 days
Database Passwords: Rotate every 180 days
Session Secrets: Rotate after security incidents
Encryption Keys: Only rotate with data migration plan

Rotating ENCRYPTION_KEY requires re-encrypting all stored OAuth tokens. Plan carefully.

Separate Environments

Use different secrets for each environment:

# Development
.env.development

# Staging
.env.staging

# Production
.env.production

Never reuse production secrets in non-production environments.

Validation & Testing

Test your configuration:

# Validate environment variables are set
docker compose config

# Test service connectivity
docker compose up -d
docker compose ps  # All services should be "healthy"

# Check logs for configuration errors
docker compose logs web
docker compose logs ai

Troubleshooting Configuration Issues

Services won't start

Check for missing required variables:

# Look for environment variable errors in logs
docker compose logs | grep -i "environment"
docker compose logs | grep -i "required"

Common issues:

Missing DATABASE_PASSWORD
Missing SESSION_SECRET or ENCRYPTION_KEY
Missing API keys for enabled features

Database connection errors

Verify database configuration:

# Check if PostgreSQL is accessible
docker compose exec web nc -zv postgres 5432

# Test database connection
docker compose exec postgres psql -U omni -d omni -c "SELECT 1;"

Check:

DATABASE_HOST matches service name
DATABASE_PASSWORD is correct
PostgreSQL is fully started

LLM provider not working

Test LLM connectivity:

# For Anthropic
curl -H "x-api-key: $ANTHROPIC_API_KEY" \
  https://api.anthropic.com/v1/messages

# For vLLM
docker compose exec web curl http://vllm:8000/health

Check:

LLM_PROVIDER is set correctly
API keys are valid
vLLM service is running (if self-hosted)

Email not sending

Test email configuration:

# Check email service logs
docker compose logs web | grep -i email

# For Resend
curl -H "Authorization: Bearer $RESEND_API_KEY" \
  https://api.resend.com/domains

Check:

EMAIL_PROVIDER is set
API key or SMTP credentials are valid
EMAIL_FROM domain is verified (for Resend)

Next Steps

Deploy with Docker

Use these variables in Docker Compose deployment

Deploy to AWS

Configure Terraform variables for AWS deployment

Resource Requirements

Understand hardware and resource needs

Security Guide

Implement security best practices

Getting Started

Deployment

Connectors

Developers

User Guide

Administration

​Quick Configuration Examples

​Database Configuration

​Redis Configuration

​Application Configuration

​Security & Encryption

​Service Ports

​Inter-Service URLs

​Feature Flags

​LLM Provider Configuration

​vLLM (Self-Hosted)

​Anthropic Claude

​AWS Bedrock

​AI Service Configuration

​Conversation Compaction

​Batch Embedding Inference (AWS Bedrock)

​Google Workspace Connector

​Slack Connector

​Atlassian Connector

​HubSpot Connector

​Filesystem Connector

​Connector Manager

​Storage Configuration

​Email Configuration

​Resend Provider

​SMTP Provider

​Search Configuration

​Logging & Monitoring

​Telemetry (OpenTelemetry)

​HTTPS & SSL (Production)

​Environment-Specific Examples

​Development (.env)

​Staging (.env)

​Production (.env)

​Security Best Practices

​Validation & Testing

​Troubleshooting Configuration Issues

​Next Steps

Deploy with Docker

Deploy to AWS

Resource Requirements

Security Guide

Quick Configuration Examples

Database Configuration

Redis Configuration

Application Configuration

Security & Encryption

Service Ports

Inter-Service URLs

Feature Flags

LLM Provider Configuration

vLLM (Self-Hosted)

Anthropic Claude

AWS Bedrock

AI Service Configuration

Conversation Compaction

Batch Embedding Inference (AWS Bedrock)

Google Workspace Connector

Slack Connector

Atlassian Connector

HubSpot Connector

Filesystem Connector

Connector Manager

Storage Configuration

Email Configuration

Resend Provider

SMTP Provider

Search Configuration

Logging & Monitoring

Telemetry (OpenTelemetry)

HTTPS & SSL (Production)

Environment-Specific Examples

Development (.env)

Staging (.env)

Production (.env)

Security Best Practices

Validation & Testing

Troubleshooting Configuration Issues

Next Steps