Agent skill

MCP Server Builder

by OneWave-AI in OneWave-AI/claude-skills

A skill your agent uses to build, wire up, or debug an MCP (Model Context Protocol) server that gives Claude access to an internal API, database, or SaaS tool.

MITAuto-check passedAgent Workflows

Install MCP Server Builder

skills CLI
$ npx skills add OneWave-AI/claude-skills --skill mcp-server-builder -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install OneWave-AI/claude-skills mcp-server-builder --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/OneWave-AI/claude-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/mcp-server-builder .claude/skills/mcp-server-builder && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
mcp-server-builder
GitHub stars
328
Token cost
~1.2k tokens
SKILL.md length
546 words
Files
1
Skills in repo
69
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses to build, wire up, or debug an MCP (Model Context Protocol) server that gives Claude access to an internal API, database, or SaaS tool.

  • Debug an MCP (Model Context Protocol) server that gives Claude access to an internal API
  • SKILL.md covers Decide the Shape First, Tool Design — the part that…, Minimum Viable Server and Security, plus 2 more sections
  • Calls claude and npx
  • The user wants Claude to connect to their own system

What it does

MCP Server Builder is an agent skill from OneWave-AI/claude-skills. Use to build, wire up, or debug an MCP (Model Context Protocol) server that gives Claude access to an internal API, database, or SaaS tool. Trigger when the user wants Claude to connect to their own system, asks for an MCP server, needs custom tools for Claude Code or the Claude app, or an existing server's tools are not showing up.

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Agent Workflows, covering MCP servers. It works with Model Context Protocol. The repository describes itself as: 200+ production-ready Claude Code skills for sales, marketing, design, engineering, and AI agent architecture. Built and maintained by OneWave AI. The licence is MIT.

When your agent uses it

  • Debug an MCP (Model Context Protocol) server that gives Claude access to an internal API
  • The user wants Claude to connect to their own system
  • Asks for an MCP server
  • Needs custom tools for Claude Code

Example prompts

  • “/mcp-server-builder”

Requirements

  • Node.js

What it can do on your machine

Read from SKILL.md and the folder at commit fc5b785. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • claude
    • npx

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npx, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

MCP Server Builder loads about 1.2k tokens when it runs. Until then it costs about 88 tokens; SKILL.md has 546 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~88
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from OneWave-AI/claude-skills at commit fc5b785, republished under its MIT licence (© OneWave-AI). 546 words, ~1,177 tokens.

Download SKILL.mdSave it as .claude/skills/mcp-server-builder/SKILL.md (or your agent's skills folder).
name
mcp-server-builder
description
Use to build, wire up, or debug an MCP (Model Context Protocol) server that gives Claude access to an internal API, database, or SaaS tool. Trigger when the user wants Claude to connect to their own system, asks for an MCP server, needs custom tools for Claude Code or the Claude app, or an existing server's tools are not showing up.
tools
Read, Write, Edit, Bash, Glob, Grep, WebFetch

MCP Server Builder

An MCP server is how Claude reaches a system it does not otherwise know about — your database, your internal API, your ops tooling. The protocol is small. Almost every real problem is a design problem: which tools to expose and what they return.

Decide the Shape First

Local (stdio) — runs on the user's machine, launched by the client, credentials from the local environment. Right for developer tooling and anything touching localhost or a local database. Only that machine gets it.

Remote (HTTP) — deployed once, reached by URL, OAuth for identity. Right when a team needs it, or when the data lives in a service rather than on a laptop. Everyone gets updates the moment you deploy.

Build remote when more than one person will use it. Retrofitting local-only to multi-user later means redoing auth from scratch.

Tool Design — the part that matters

Tools are an interface for a model, not a REST API. The common failure is exposing twenty thin endpoint wrappers and watching the model chain them badly.

  • Name the job, not the endpoint. get_pipeline_board beats query_deals plus query_stages plus query_owners.
  • One call should answer a real question. If using your server always takes three calls in the same order, that sequence should have been one tool.
  • Descriptions are prompts. Say what the tool is for, when to reach for it, and what it does not do. This text is the only instruction the model gets.
  • Return prose-shaped, compact results. Trim nulls and internal ids nobody will use. A 40KB JSON blob buries the answer and eats the window.
  • Paginate and cap. Always a default limit. Say in the response when results were truncated, and how to get the rest.
  • Separate reads from writes. Writes get explicit, narrow tools with clear names — never a generic execute that takes arbitrary input.
  • Errors are instructions. "No company matched 'acme' — try search_companies first" gets recovered from; "500 Internal Server Error" does not.

Ten well-shaped tools beat fifty thin ones.

Show full SKILL.md (216 more words)Show less

Minimum Viable Server

ts
import { McpServer } from '@modelcontextprotocol/sdk/server/mcp.js'
import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js'
import { z } from 'zod'

const server = new McpServer({ name: 'acme-crm', version: '1.0.0' })

server.tool(
  'get_deal',
  'Fetch one deal with its owner, stage, and recent activity. Use when the user names a specific deal or company.',
  { id: z.string().describe('Deal id or display id like D-1042') },
  async ({ id }) => ({ content: [{ type: 'text', text: await renderDeal(id) }] }),
)

await server.connect(new StdioServerTransport())

Check the SDK's current API before writing more than this — the interface has changed across releases, and code written from memory tends to target an older one.

Security

  • Least privilege on the credential. A read-only token for a read-only server.
  • Never take a raw query string from the model into a database. Parameterize, allow-list tables, scope by tenant on the server side.
  • Scope every request to the authenticated user. A remote server that trusts a user id in the arguments is an authorization hole.
  • Never log or return secrets. Redact tokens in error paths.
  • Rate-limit writes.

Debugging

bash
npx @modelcontextprotocol/inspector node build/index.js   # exercise tools directly
claude mcp list                                           # is it registered
claude mcp get <name>                                     # config and status

Tools missing in the client is nearly always one of: the server crashed on startup (run it by hand and read stderr), a stdio server wrote a stray console.log to stdout and corrupted the protocol stream, the config points at a stale build, or the client was not restarted. For stdio servers, all logging goes to stderr — stdout belongs to the protocol.

Ship Checklist

  • Every tool description says when to use it and when not to.
  • Every tool tested through the inspector, not only through the client.
  • Results capped and truncation stated.
  • Errors say what to try next.
  • Credentials from the environment, never in the repo.
  • README: install, config snippet, and one example prompt per tool.

© OneWave-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in mcp-server-builder of OneWave-AI/claude-skills.

Open the folder on GitHubat commit fc5b785

Compare with similar skills

MCP Server Builder next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

MCP Server Builder compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
MCP Server Builder this skillOneWave-AI/claude-skills328—~1.2kAutomated safety check: PassMIT
MCP Server Builderanthropics/skills180k63 repos~2.3kAutomated safety check: PassApache-2.0
MCP Server BuildershareAI-lab/learn-claude-code78k5 repos~1.2kAutomated safety check: PassMIT
MCP Integration for Pluginsanthropics/claude-plugins-official38k11 repos~3.1kAutomated safety check: PassApache-2.0
Crush Configurationcharmbracelet/crush29k—~3.7kAutomated safety check: PassCustom licence
Context Mode Output Sandboxmksglu/context-mode26k—~4.1kAutomated safety check: PassCustom licence

Similar skills

  • MCP Server Builder

    anthropics/skills

    Official

    Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.

    180k GitHub starsUsed in 63 repos~2.3k tokens
    Agent WorkflowsAuto-check passed
  • MCP Server Builder

    shareAI-lab/learn-claude-code

    Walks through building MCP servers in Python or TypeScript that expose tools, resources and prompts to Claude, with templates, registration and testing.

    78k GitHub starsUsed in 5 repos~1.2k tokens
    Agent WorkflowsAuto-check passed
  • MCP Integration for Plugins

    anthropics/claude-plugins-official

    Official

    Explains how to bundle Model Context Protocol servers in a Claude Code plugin, covering config files, stdio, SSE, HTTP and WebSocket server types, and authentication.

    38k GitHub starsUsed in 11 repos~3.1k tokens
    Agent WorkflowsAuto-check passed
  • Crush Configuration

    charmbracelet/crush

    Explains how to configure the Crush coding agent with crushrc or crush.json, covering providers, models, LSPs, MCP servers, hooks, permissions and config precedence.

    29k GitHub stars~3.7k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Context Mode Output Sandbox

    mksglu/context-mode

    Routes large command, file, API and browser output through context-mode tools so only the needed result enters the agent's context, instead of dumping it via Bash.

    26k GitHub stars~4.1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Migrates the compatible subset of settings and global file-based MCP servers from the Warp desktop app into Warp Agent CLI without exposing credentials or state.

    65k GitHub starsUsed in 1 repo~2.1k tokens
    Agent WorkflowsAuto-check passed

More from OneWave-AI/claude-skills

All 69 skills in this repo
  • CRM Data Cleanup

    OneWave-AI/claude-skills

    Finds duplicate and junk records in a CRM CSV export with fuzzy matching, normalizes fields and writes a reviewable merge plan plus import-ready files without touching the live CRM.

    328 GitHub stars~2.5k tokensUpdated 7 days ago
    Auto-check passed
  • Design Export Repair

    OneWave-AI/claude-skills

    Repairs broken decks and PDFs exported from Claude Design or similar AI deck generators: clipped text, wrong fonts and corrupted .pptx package structure.

    328 GitHub stars~2.6k tokensUpdated 7 days ago
    Auto-check passed
  • Bi Measure Builder

    OneWave-AI/claude-skills

    Writes, explains, debugs, and optimizes BI calculations - Power BI / Fabric DAX measures and calculated columns, Tableau calculated fields (FIXED/INCLUDE/EXCLUDE LOD expressions, table…

    328 GitHub stars~2.2k tokensUpdated 7 days ago
    Auto-check passed
  • Bookkeeping Close

    OneWave-AI/claude-skills

    Categorizes transactions, reconciles bank and card statements to the ledger, works a month-end checklist and prepares a close package, without ever forcing a balance.

    328 GitHub stars~2k tokensUpdated 7 days ago
    Auto-check passed
  • CSV and Excel Merger

    OneWave-AI/claude-skills

    Combines CSV, TSV and Excel files into one verified table with pandas, by stacking or joining, mapping columns, normalizing keys and removing duplicates.

    328 GitHub stars~1.6k tokensUpdated 7 days ago
    Auto-check passed
  • Sec Filing Puller

    OneWave-AI/claude-skills

    Pulls financial statement numbers for US public companies straight from SEC EDGAR's free official XBRL APIs (companyfacts, companyconcept, frames, submissions) into a cited table.

    328 GitHub stars~2.1k tokensUpdated 7 days ago
    Auto-check passed

Categories

Questions about MCP Server Builder

What does MCP Server Builder do?

A skill your agent uses to build, wire up, or debug an MCP (Model Context Protocol) server that gives Claude access to an internal API, database, or SaaS tool. MCP Server Builder is an agent skill from OneWave-AI/claude-skills. Use to build, wire up, or debug an MCP (Model Context Protocol) server that gives Claude access to an internal API, database, or SaaS tool.

When should I use MCP Server Builder?

MCP Server Builder fits situations like: debug an MCP (Model Context Protocol) server that gives Claude access to an internal API; the user wants Claude to connect to their own system; asks for an MCP server; needs custom tools for Claude Code.

How do I install MCP Server Builder in Claude Code?

Run `npx skills add OneWave-AI/claude-skills --skill mcp-server-builder -a claude-code`. Or copy the skill folder (mcp-server-builder in OneWave-AI/claude-skills) into .claude/skills/mcp-server-builder in your project. Claude Code loads it when a task matches its description.

How do I install MCP Server Builder in Codex?

Run `npx skills add OneWave-AI/claude-skills --skill mcp-server-builder -a codex`. Or copy the skill folder (mcp-server-builder in OneWave-AI/claude-skills) into .agents/skills/mcp-server-builder in your project. Codex loads it when a task matches its description.

Can I use MCP Server Builder in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add OneWave-AI/claude-skills --skill mcp-server-builder -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/mcp-server-builder, .gemini/skills/mcp-server-builder, .github/skills/mcp-server-builder and .opencode/skills/mcp-server-builder in your project.

What does MCP Server Builder need to run?

Going by SKILL.md and its folder, MCP Server Builder needs the command-line tools its instructions call (claude and npx). Our summary lists: Node.js.

Does MCP Server Builder access the network?

SKILL.md contains no URLs. Its commands use npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is MCP Server Builder safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does MCP Server Builder use?

MCP Server Builder is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does MCP Server Builder use?

About 1.2k tokens (SKILL.md is roughly 4.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to MCP Server Builder?

Skills that share tags, products or a category with MCP Server Builder: MCP Server Builder (anthropics/skills, 180k stars), MCP Server Builder (shareAI-lab/learn-claude-code, 78k stars), MCP Integration for Plugins (anthropics/claude-plugins-official, 38k stars) and Crush Configuration (charmbracelet/crush, 29k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains MCP Server Builder?

OneWave-AI (a GitHub organization) maintains it in OneWave-AI/claude-skills, which has 328 GitHub stars. The repository holds 69 skills in this directory. The repository was last updated on October 2, 2026.

Source: OneWave-AI/claude-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.