Deployment

Strategies and recommendations for deploying Echo Agent in production.

---

Deployment Checklist

• [ ] Choose appropriate security profile
• [ ] Configure model provider API keys
• [ ] Set up Gateway authentication tokens
• [ ] Configure TLS via reverse proxy
• [ ] Set up backup schedule
• [ ] Configure monitoring/alerting
• [ ] Review tool permissions

Single-Server Deployment

The simplest production setup:

pip install "echo-agent[all]"
echo-agent setup
echo-agent gateway install
echo-agent gateway start

Reverse Proxy

For public-facing deployments, always use a reverse proxy with TLS:

• Nginx configuration
• Caddy configuration

Public Exposure Warning

Gateway listens on 127.0.0.1 by default. If you need external access, you must configure:

1. A reverse proxy (Nginx/Caddy) with TLS
2. gateway.auth.mode: allowlist with strong random api_tokens / admin_tokens
3. gateway.auth.allowed_origins and allowed_hosts restricted to the real domain
4. security.profile: public_gateway, with tools.profile reduced as needed
5. Source-network and rate limits at the firewall or reverse-proxy layer

See Security Hardening and Gateway Authentication.

Resource Requirements

Scale	RAM	CPU	Disk
Personal (1-2 channels)	512MB	1 core	1GB
Small team (3-5 channels)	1GB	2 cores	5GB
Heavy usage (all channels + knowledge)	2GB+	4 cores	10GB+

Note

Model inference latency is typically the bottleneck, not local compute.

Environment Variables

Set secrets via environment variables rather than config files:

export ECHO_AGENT_MODELS__PROVIDERS__0__API_KEY="sk-..."
export ECHO_AGENT_GATEWAY__AUTH__API_TOKENS__0="your-token"

See Environment Variables Reference.