MCP Servers

Built-in MCP servers that open your workspace to every AI tool

Every app, workflow, and integration in your Runwork workspace is automatically available via MCP (Model Context Protocol). Connect Claude Desktop, Cursor, Codex, or any MCP-compatible tool and they can read your data, trigger your workflows, call your endpoints, and use your skills. One workspace, every AI tool.

MCP Connections

3 connected

Workspace MCP Server Active

https://acme.runwork.ai/mcp

Connected AI Tools

Claude Desktop Workspace access

Connected

Cursor CRM app only

Connected

Custom AI Agent Workspace access

Connected

Available MCP Tools

query_entity create_entity call_endpoint trigger_workflow trigger_schedule call_integration send_message list_skills +14 more

22 tools available

Access from anywhere

AI Agents

Ask your AI agent:

Connect your AI agent

Web Dashboard

Live health checks, tool explorer, quick-connect with copyable endpoints. Browse all servers across four tabs: All, App MCPs, External Servers, Integrations.

Open in dashboard

CLI

Capabilities

Workspace MCP Server

One connection point exposes your entire workspace. Every app, entity, workflow, schedule, integration, and skill, all available through a single MCP endpoint.

Per-App MCP Servers

Each app gets its own MCP server URL. Not just monitoring: teams and AI tools can manage data, trigger workflows, run schedules, upload and download files, all through the app's MCP server. Full read and write access.

Full Tool Mapping

Everything maps to MCP tools: query entities, call endpoints, trigger workflows, run schedules, use integrations, send channel messages, read skills. Your full workspace as AI tools.

Resources & Prompts

Entity schemas, app manifests, and skill content served as MCP resources and prompts. External AI tools get rich context about your business, not just raw tool calls.

API Key Authentication

Secure access with workspace API keys. Scope keys to specific apps. Create, rotate, and revoke keys from your dashboard. Same security model as your APIs.

Standard Protocol

Streamable HTTP and SSE transports. Works with any tool that speaks MCP. No custom SDKs, no vendor lock-in.

MCP Servers Hub

A dedicated dashboard to manage all MCP servers. Live health checks with latency and status. A tool explorer to browse and test available tools. Quick-connect with copyable endpoint URLs and inline API key creation.

Tool Explorer

Browse every tool exposed by any MCP server. See tool names, descriptions, and schemas. Test tools directly from the dashboard with parameter inputs and live results.

Use Cases

Claude Desktop + your data Cursor with business context Custom AI agents Cross-tool orchestration Automated reporting Data enrichment pipelines

Why It Matters

Your workspace is instantly available to Claude, Cursor, and any MCP client
No setup per tool: connect once, access everything
Per-app servers for granular access control
Open protocol: not locked into any single AI vendor
Every app is a fully functional MCP server with read and write access, not just monitoring

How It Works

MCP (Model Context Protocol) is the open standard for connecting AI tools to data sources. Runwork gives every workspace a built-in MCP server, so your apps, data, and workflows are accessible to any AI tool that speaks MCP.

The workspace MCP server exposes everything: entities (customers, orders, products), endpoints, workflows, schedules, integrations, components, agents, file storage, channels, and skills. All mapped to MCP tools with proper schemas and descriptions. Connect Claude Desktop and ask it to "query all customers who signed up this month" or "trigger the weekly report workflow."

Every app is automatically an MCP server

This is automatic. You don't configure anything. The moment you deploy an app, it gets its own MCP server URL. Add an entity to your app? It's immediately queryable via MCP. Add a workflow? It's triggerable. Connect an integration? It's callable. Every feature you add to your app is instantly available as an MCP tool.

Per-app MCP servers give focused access. Connect Cursor to just your CRM app and it gets CRM tools without seeing your entire workspace. Connect Claude Desktop to your Inventory Tracker and it can query stock levels, trigger reorder workflows, and download reports. Each app's MCP server exposes that app's entities, endpoints, workflows, schedules, integrations, file storage, and agents. Full read and write access, not just monitoring.

This is what makes Runwork's AI ecosystem compound. Your team creates 10 apps. That's 10 MCP servers, each with its own tools, plus the workspace server that aggregates everything. Any AI tool your team uses, Claude Code, Cursor, Codex, Claude Desktop, gets access to all of it. The more apps you create, the more your AI tools can do.

External MCP servers and integrations

It's not just your apps. Connected integrations also appear as MCP tools. Connect HubSpot, and your AI agents get a hubspot_api tool. Connect Stripe, and they can query payments and subscriptions. These integration tools show up alongside your app tools in the MCP Servers Hub.

You can also bring in external MCP servers. Add any MCP-compatible server to your workspace and it becomes available to your team's AI tools. The Hub shows live health status for each server: connected, slow, unreachable, or error, with latency measurements. The Tool Explorer lets you browse and test any tool from any server directly in the dashboard.

Authentication and protocol

Authentication uses your existing workspace API keys. Keys can be scoped to specific apps for per-app MCP access. The same security model you use for your public APIs applies to MCP connections. Both Streamable HTTP and SSE transports are supported. Any MCP-compliant client works out of the box.

MCP also exposes resources (entity schemas, app manifests) and prompts (app skills, channel instructions). External tools don't just get tool calls. They get the full context of your business. The same skills your workspace AI uses internally are available to external tools too.

MCP during local development

MCP servers connect naturally to local development. When you develop with the Runwork CLI, your local AI tools can connect to your app's MCP server while you code. Claude Code reads SKILL.md (auto-generated for every app, also downloadable from the dashboard) and CLAUDE.md + AGENTS.md (created on the fly by the platform during runwork dev) for framework knowledge, and connects via MCP for live data access. This means your AI assistant can query real data, test workflows, and verify integrations while building features.

The MCP Servers Hub is a dedicated dashboard for managing all MCP connections. Four tabs organize your servers: All, App MCPs, External Servers, and Integrations. Each external server shows live health status with connection latency, tool count, and color-coded indicators (connected, slow, unreachable, error). The Tool Explorer lets you browse every tool exposed by any server and test them directly with parameter inputs and live results. Quick-connect shows copyable endpoint URLs with inline API key creation, so connecting a new AI tool takes seconds. Connected integrations appear as MCP tools too, so your HubSpot, Stripe, and Slack connections are browsable alongside app-specific tools.

Frequently Asked Questions

What is MCP (Model Context Protocol)?

MCP is an open standard for connecting AI tools to data sources and capabilities. It defines how AI tools discover available tools, call them, and receive results. Think of it as a universal adapter between AI assistants and your business data. Runwork implements MCP so your workspace is accessible to any compatible AI tool.

Which AI tools work with Runwork MCP?

Any tool that supports the MCP protocol. This includes Claude Desktop, Cursor, Codex, and a growing ecosystem of AI tools. Both Streamable HTTP and SSE transports are supported, so any MCP-compliant client can connect to your Runwork workspace.

What is the difference between workspace and per-app MCP servers?

The workspace MCP server exposes everything across all your apps through one connection. Per-app MCP servers expose only a single app's capabilities. Use workspace MCP when you want broad access, and per-app MCP when you want focused, scoped access. Per-app servers are also useful for sharing a specific app with an external tool without exposing your full workspace.

How do I connect Claude Desktop to my workspace?

Add your workspace MCP server URL and API key to Claude Desktop's MCP configuration. Claude will automatically discover all available tools, resources, and prompts. You can then ask Claude to interact with your workspace data, trigger workflows, and use your app skills. Per-app URLs are also available if you want to connect to a specific app.

Is MCP access secure?

Yes. All MCP connections are authenticated with workspace API keys. Keys can be scoped to specific apps, so you control exactly what each connection can access. You can create, rotate, and revoke keys from your workspace dashboard. The same security model that protects your public APIs applies to MCP connections.

Related Features

APIs Skills Integrations CLI

See How Teams Use MCP Servers

AI Ecosystem CTOs SaaS Teams

Ready to try MCP Servers?

Try the work cloud for AI agents.

Get Started Explore More Features