Environment Variables

Configure your S3 Documentation MCP server using environment variables in your .env file.

Quick Setup

cp env.example .env

Then edit .env with your specific configuration.

Essential Variables

S3 Configuration

# Your S3 bucket name
S3_BUCKET_NAME=your-bucket-name

# S3 access credentials
S3_ACCESS_KEY_ID=your-access-key
S3_SECRET_ACCESS_KEY=your-secret-key

# AWS region (e.g., us-east-1, eu-west-1)
S3_REGION=us-east-1

# Optional: Custom S3 endpoint for non-AWS providers
# Leave empty for AWS S3
S3_ENDPOINT=

Examples for Different Providers

AWS S3:

S3_ENDPOINT=
S3_REGION=us-east-1

MinIO:

S3_ENDPOINT=https://minio.example.com
S3_REGION=us-east-1  # Can be any value

Scaleway:

S3_ENDPOINT=https://s3.fr-par.scw.cloud
S3_REGION=fr-par

Cloudflare R2:

S3_ENDPOINT=https://<account-id>.r2.cloudflarestorage.com
S3_REGION=auto

Embedding Provider

Choose your embedding provider:

# ollama (default) or openai
EMBEDDING_PROVIDER=ollama

For Ollama (Local):

OLLAMA_BASE_URL=http://localhost:11434
OLLAMA_EMBEDDING_MODEL=nomic-embed-text

For OpenAI (Cloud):

OPENAI_API_KEY=sk-...your-key...
OPENAI_EMBEDDING_MODEL=text-embedding-3-small  # or text-embedding-3-large

See Embedding Providers for detailed comparison.

Optional Variables

Server Configuration

# Server port (default: 3000)
PORT=3000

# Server host (default: 0.0.0.0)
HOST=0.0.0.0

RAG Configuration

# Number of results returned by search (default: 4)
RAG_MAX_RESULTS=4

# Chunk size for document splitting (default: 1000)
RAG_CHUNK_SIZE=1000

# Overlap between chunks (default: 200)
RAG_CHUNK_OVERLAP=200

Synchronization

# Sync mode: startup (default), periodic, or manual
SYNC_MODE=startup

# For periodic mode: interval in minutes (default: 60)
SYNC_INTERVAL_MINUTES=60

See Synchronization Modes for details.

Security

# Enable API key authentication (default: false)
ENABLE_AUTH=false

# API key for authentication (required if ENABLE_AUTH=true)
MCP_API_KEY=your-secret-key-here

See Security & Authentication for details.

Vector Store

# Path to store vector indices (default: ./data/hnswlib-store)
VECTOR_STORE_PATH=./data/hnswlib-store

Complete Example

Here’s a complete .env file example:

# S3 Configuration
S3_BUCKET_NAME=my-docs-bucket
S3_ACCESS_KEY_ID=AKIAIOSFODNN7EXAMPLE
S3_SECRET_ACCESS_KEY=wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
S3_REGION=us-east-1
S3_ENDPOINT=

# Embedding Provider (Ollama - Local)
EMBEDDING_PROVIDER=ollama
OLLAMA_BASE_URL=http://localhost:11434
OLLAMA_EMBEDDING_MODEL=nomic-embed-text

# Server
PORT=3000
HOST=0.0.0.0

# RAG Configuration
RAG_MAX_RESULTS=4
RAG_CHUNK_SIZE=1000
RAG_CHUNK_OVERLAP=200

# Synchronization
SYNC_MODE=startup
SYNC_INTERVAL_MINUTES=60

# Security (disabled by default)
ENABLE_AUTH=false
# MCP_API_KEY=

# Vector Store
VECTOR_STORE_PATH=./data/hnswlib-store

Environment-Specific Configuration

Development

EMBEDDING_PROVIDER=ollama
SYNC_MODE=manual
ENABLE_AUTH=false

Production

EMBEDDING_PROVIDER=openai
SYNC_MODE=periodic
SYNC_INTERVAL_MINUTES=30
ENABLE_AUTH=true
MCP_API_KEY=<strong-random-key>

Validation

The server validates your configuration on startup and will:

✅ Show clear error messages for missing required variables
✅ Warn about invalid values
✅ Fall back to safe defaults where possible
✅ Test S3 connectivity on startup