lorenzespinosa/openclaw-n8n-stack
Production-ready Docker stack: OpenClaw AI agent + n8n workflow automation + PostgreSQL. One-command deploy, pre-wired MCP, multi-agent config, token savings.
Platform-specific configuration:
{
"mcpServers": {
"openclaw-n8n-stack": {
"command": "npx",
"args": [
"-y",
"openclaw-n8n-stack"
]
}
}
}Add the config above to .claude/settings.json under the mcpServers key.
[](LICENSE) [](docker-compose.yml)
> Production-ready Docker stack pairing OpenClaw (AI agent gateway) with n8n (workflow automation). One command to deploy, pre-wired MCP, pre-configured multi-agent.
---
graph TB
User([User / Slack]) -->|Chat| OC[OpenClaw Gateway<br/>:18789]
subgraph Docker Network
OC -->|MCP over SSE| N8N[n8n<br/>:5678]
N8N -->|SQL| PG[(PostgreSQL<br/>:5432)]
end
OC -->|API| OR[OpenRouter / Anthropic]
N8N -->|Webhooks| EXT[External APIs<br/>Lawmatics · Filevine<br/>Clio · OpenPhone · Airtable]
subgraph Agents
ORCH[Orchestrator] -->|spawn| OPS[Ops Specialist]
ORCH -->|spawn| RES[Research Specialist]
end
OC --- Agents
style OC fill:#6366f1,color:#fff
style N8N fill:#ff6d5a,color:#fff
style PG fill:#336791,color:#fffHow it works:
---
# 1. Clone
git clone https://github.com/lorenzespinosa/openclaw-n8n-stack.git
cd openclaw-n8n-stack
# 2. Configure
cp .env.example .env
# Edit .env — set POSTGRES_PASSWORD, N8N_ENCRYPTION_KEY,
# OPENCLAW_GATEWAY_TOKEN, and at least one AI provider key
# 3. Deploy
docker compose up -d
# 4. Verify
docker compose ps # All 3 services should be "healthy"
curl http://localhost:5678/healthz # n8n
curl http://localhost:18789/healthz # OpenClawAccess points: | Service | URL | Purpose | |---------|-----|---------| | n8n | http://localhost:5678 | Workflow editor & webhook endpoints |
Loading reviews...