@@ -128,7 +128,7 @@ As of October 2026, each product fits a different center of gravity: Sim for vis
 | Self-hosting | Yes | [Yes, subject to n8n's license terms](https://docs.n8n.io/deploy/host-n8n) | Depends on the component; SDK application code can run in your environment, while managed platform services remain OpenAI services |
 | Model strategy | Multiple model providers, workspace BYOK, and local models on self-hosted deployments | [Multiple chat-model nodes](https://docs.n8n.io/build/integrate-ai/langchain-in-n8n) alongside a broad automation ecosystem | Designed around OpenAI models and services |
 | Human approval | Human in the Loop pauses a run and returns form fields for downstream routing | [Human review can pause an AI Agent before selected tool calls](https://docs.n8n.io/build/integrate-ai/ai-examples/human-in-the-loop-for-tools) | [Agents SDK runs can pause for approval decisions](https://developers.openai.com/api/docs/guides/agents/guardrails-approvals) |
-| Application connectivity | Built-in integrations, APIs, webhooks, and MCP support | Built-in and community nodes | Tools, MCP support, and custom application code |
+| Application connectivity | Built-in integrations, APIs, webhooks, and MCP (Model Context Protocol), a standard that lets agents connect to external tools and data | Built-in and community nodes | Tools, MCP support, and custom application code |
 | Local models | Ollama, vLLM, LM Studio, and LiteLLM on self-hosted deployments | [Ollama Chat Model node](https://docs.n8n.io/integrations/builtin/cluster-nodes/sub-nodes/n8n-nodes-langchain.lmchatollama) | Not the focus of the OpenAI-hosted stack |
 | Billing basis | Cloud plan plus model usage, with workspace BYOK on every cloud plan | [Cloud pricing based on monthly workflow executions](https://n8n.io/pricing/) | [Model tokens and billable tool usage](https://openai.com/api/pricing/) |
 | Main advantage | Agent-first design with permissive core code ownership | Integration-first automation | Tight OpenAI platform integration |