SaaS Factory Joins the MCP Ecosystem — Bidirectional AI Tool Interoperability
SaaS Factory Joins the MCP Ecosystem
Bidirectional AI Tool Interoperability via Model Context Protocol
Released in v1.0.131
Model Context Protocol (MCP) is quickly becoming the standard language that AI agents use to talk to tools. Cursor has an MCP marketplace. Devin has an MCP marketplace. Claude Desktop supports MCP connectors natively. And as of v1.0.131, SaaS Factory does too — on both sides of the protocol.
What is MCP?
Model Context Protocol is an open standard that defines how AI agents discover and invoke external tools. Rather than every AI platform building custom integrations to every service, MCP provides a single, consistent interface:
- A server exposes a set of tools with typed inputs and outputs.
- A client (agent or IDE) discovers those tools and calls them using the protocol.
- Any client can call any server. Any server can be called by any client.
Think of it as a universal adapter for AI tool interoperability.
What SaaS Factory Ships in v1.0.131
SaaS Factory as an MCP Server
SaaS Factory now runs an MCP server that exposes its platform capabilities as callable tools. Any MCP-compatible client can connect and invoke SaaS Factory's agents directly — no custom API wiring required.
What this enables:
- Invoke SaaS Factory agents from Cursor while writing code.
- Trigger factory pipelines from Claude Desktop in a conversational interface.
- Call SaaS Factory tools from Devin or any other MCP-compatible AI agent.
- Build your own MCP clients that orchestrate SaaS Factory alongside other services.
SaaS Factory as an MCP Consumer
SaaS Factory's internal agents can now connect to external MCP servers. Instead of the platform maintaining a growing library of hand-rolled integrations, agents can discover and call any MCP-compatible tool at runtime.
Examples of what agents can now tap into:
- Stripe MCP — billing, subscription, and payment operations without a custom Stripe client.
- GitHub MCP — repository management, PR operations, and code search.
- Linear MCP — issue tracking and project management.
- Any other MCP server published to the ecosystem, today or in the future.
Before and After
| Before v1.0.131 | After v1.0.131 | |
|---|---|---|
| Calling SaaS Factory from Cursor | ❌ Not possible | ✅ Connect via MCP |
| Calling SaaS Factory from Claude Desktop | ❌ Not possible | ✅ Connect via MCP |
| Agents using Stripe capabilities | Custom integration required | ✅ Stripe MCP |
| Agents using GitHub capabilities | Custom integration required | ✅ GitHub MCP |
| Adding a new external tool | Write and maintain custom code | ✅ Connect any MCP server |
The Bigger Picture
SaaS Factory's mission is to orchestrate 21+ specialised agents that never stop improving your product. MCP multiplies the reach of every one of those agents. Rather than each agent being limited to the integrations SaaS Factory has manually built, they now have access to the entire MCP ecosystem — and that ecosystem grows every week.
At the same time, exposing SaaS Factory itself as an MCP server means the platform becomes a first-class citizen in every AI-native workflow. Wherever your team works — Cursor, Claude, Devin, or a custom agent — SaaS Factory's capabilities are one MCP connection away.
Getting Started
To connect an MCP client to your SaaS Factory instance, point it at your MCP server endpoint and authenticate with your API credentials. Consult the API Reference for endpoint details and available tools.
To configure your SaaS Factory agents to consume an external MCP server, add the server URL and credentials to your agent configuration. Agents will automatically discover the available tools at runtime.