# When should you build an MCP server instead of a skill?

Canonical URL: https://bestskillsforclaude.com/en/skills/mcp-builder
Language: en
Skill: mcp-builder
Publisher: Anthropic (https://github.com/anthropics)
Source: https://github.com/anthropics/skills/tree/main/skills/mcp-builder
License as stated by the publisher: Apache 2.0
Runtime: Claude apps and Claude Code
Category: Extend Claude
Last reviewed: 2026-08-05

When the agent needs to do something, not to know something.

## Short answer
The mcp-builder skill helps you design and implement Model Context Protocol servers, which is how an agent gets real capabilities such as querying your database or calling your internal API. Choose a server when the agent needs to act on a system, and a skill when it needs to follow a procedure.

Best for: Exposing internal systems, APIs and data sources to agents in a controlled way.
Skip it if: What you need is procedural guidance, which is a skill and far cheaper to maintain.

## What to remember
- Tool design is the whole game. A poorly named tool is an unused tool.
- Fewer, well scoped tools beat many overlapping ones.
- Error messages are part of the interface, because the agent reads them and decides what to do next.
- Servers grant capability, and capability needs authorisation boundaries.

## What it changes in practice
The difficulty in MCP work is not the protocol, it is the interface design. An agent picks tools from names and descriptions, so a tool called process that takes an options object will be misused constantly.
Building with explicit guidance keeps the surface small and legible, which is what makes the difference between a server the agent uses correctly and one it avoids or misapplies.

## Where it disappoints
It does not decide your security model. What the agent may read, write and delete is a decision that belongs to you and needs enforcing on the server side.
Protocol details also move. Check the current specification rather than assuming the skill's snapshot is up to date.

## Install and first run
1. Install it from anthropics/skills under Apache 2.0.
2. Design the tool surface first: few tools, clear names, obvious arguments.
3. Test with the vaguest realistic request, because that is what exposes bad tool boundaries.

## Questions
### Skill or MCP server?
Skill for procedure and conventions. Server for capability, meaning access to a system the agent cannot otherwise reach.

### What is the most common design mistake?
Too many overlapping tools with vague names, which makes selection unreliable.

### Do error messages matter?
Yes. The agent reads them and chooses its next step, so treat them as part of the interface.

## Sources
- anthropics/skills on GitHub: https://github.com/anthropics/skills
- anthropics/skills, mcp-builder skill directory: https://github.com/anthropics/skills/tree/main/skills/mcp-builder

## Other languages
- Spanish: https://bestskillsforclaude.com/es/skills/mcp-builder
- Brazilian Portuguese: https://bestskillsforclaude.com/pt-br/skills/mcp-builder
- Japanese: https://bestskillsforclaude.com/ja/skills/mcp-builder