PLAYBOOK · AUTOMATION
Make n8n and CrewAI first-class tools inside VDF AI Networks.
You already have automation: n8n flows, CrewAI crews, internal Python jobs. Don't replace them — let VDF AI orchestrate them. Register each workflow as a Custom HTTP tool; VDF AI exposes it via the internal MCP server, discovers its capabilities, and composes it with native agents.
The problem
Workflows and agent platforms live in silos
n8n knows how to fetch and post. CrewAI knows how to plan. Neither was built to orchestrate the other, or to be governed by a self-evolving router that sees energy, cost, and capability across the whole stack.
The VDF AI approach
One MCP plane over all your automation
Register each n8n webhook or CrewAI crew as a typed HTTP tool. VDF AI's Agent Hub serves it through its internal MCP server (port 7001) alongside the 44 built-in tools. Networks pick it up like any other capability.
REFERENCE ARCHITECTURE
Workflows become tools become networks
Webhook endpoints
tool_catalog · type=http
Agent Hub :7001
HTTP-exposed
parameters_schema
Capability discovery + SEEMR routing
PLAYBOOK · STEP BY STEP
From local automation to orchestrated capability
Expose your n8n flow as a webhook
Add a Webhook trigger to your n8n flow. Test the URL, capture the JSON input/output shape — that becomes your parameters schema.
Wrap CrewAI behind a tiny FastAPI
CrewAI runs as Python. A 30-line FastAPI wrapper exposes POST /crew/run. That's all VDF AI needs to call it.
Register both as Custom HTTP Tools
POST /api/tools/http
{
"tool_name": "n8n_invoice_extract",
"endpoint_url": "https://n8n.internal/webhook/invoice",
"http_method": "POST",
"parameters_schema": { "type":"object", "properties": { "file_url": {"type":"string"} }, "required": ["file_url"] }
}Agent Hub stores the tool with tool_type='http' and merges it into GET /api/tools.
Let VDF AI discover capabilities
Networks v3 reads each tool's schema and description, then ranks it during intent decomposition. Your workflows become candidates the planner can pick from automatically.
Orchestrate with a Network
Build a Network in Network Labs that combines built-in MCP tools, your n8n flows, your CrewAI crews, and VDF agents. SEEMR learns which tool produces the best result for each sub-intent.
OUTCOMES
Keep what works, orchestrate everything
workflows rewritten — your n8n and CrewAI investments stay intact.
orchestration plane across legacy automation and modern agents.
learns which workflow wins which sub-task — no manual routing rules.
SEEMR REFERENCE
Self-evolving routing across every tool
Whether the next step calls a built-in MCP, an n8n flow, or a CrewAI crew, SEEMR picks the one that matches the run's cost, latency, and energy constraints.
GET IN TOUCH
You Have Questions
Tell us what you’re trying to achieve—governed AI Networks, enterprise RAG, deep integrations, or on‑premise deployment. We’ll help you map the right architecture, security posture, and rollout path. If you’re moving beyond AI pilots and need scalable, auditable execution, reach out—our team is ready to help.