Extend agent capabilities with Model Context Protocol servers — additional tools and integrations.
MCP (Model Context Protocol) servers extend what agents can do by providing additional tools. They follow an open standard for connecting AI models to external capabilities.
An MCP server exposes tools that agents can call. For example:
To add or manage MCP server configurations:
| Field | Description |
|---|---|
| Name | Display name for this server |
| Command | The executable command (e.g., node, python) |
| Args | Command arguments (e.g., path to the MCP server script) |
| Env | Environment variables the server needs |
Both extend agent capabilities, but differently:
| Skills | MCP Servers | |
|---|---|---|
| Format | Markdown instructions + scripts | Running process with tool API |
| How agents use them | Read instructions, follow steps | Call tools directly |
| Best for | Complex workflows, multi-step processes | Direct tool access, API integrations |
| Setup | Content in database | Server process configuration |