create-mcp-servers — Claude Code skill: what it does, how to install
create-mcp-servers guides development of Model Context Protocol servers in Python and TypeScript for Claude Code, covering architecture, secrets isolation, and debugging.
MIT ★ 1,988 updated 2026-04-01
/plugin marketplace add glittercowboy/taches-cc-resources
/plugin install taches-cc-resources@taches-cc-resourcesWhat create-mcp-servers does
The skill runs step-by-step workflows to scaffold and verify MCP servers. It determines the architecture based on operation count. If the server handles one or two operations, it generates flat tools. When scaling to three or more endpoints, it applies an on-demand discovery model with four meta-tools and an operations.json map.
Under the hood, it includes project init scripts, starter templates for Python and TypeScript, and documentation for OAuth isolation. For Python setups, uv is mandatory instead of standard pip.
Without absolute binary paths discovered via which uv or which node, the Claude Code stdio transport fails to launch the process.
What it helps with
- Scaffold a Python or TypeScript MCP server from scratch
- Structure large APIs using an on-demand tool discovery pattern
- Configure stdio transport with environment variable expansion
- Troubleshoot connection failures in Claude Code settings
Example prompts for the agent
- Create a new TypeScript MCP server to fetch metrics from an external API
- Migrate an existing Python MCP server to the on-demand discovery pattern
- Troubleshoot why my server fails to appear in claude mcp list
How to install create-mcp-servers
1Plugin marketplace
/plugin marketplace add glittercowboy/taches-cc-resources/plugin install taches-cc-resources@taches-cc-resourcesThis installs the whole taches-cc-resources plugin with all its skills
2Skills CLI (Vercel)
npx skills add glittercowboy/taches-cc-resources --skill create-mcp-servers -a claude-code3Manually: put the files in place
~/.claude/skills/create-mcp-servers/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/create-mcp-servers/
1Skills CLI (Vercel)
npx skills add glittercowboy/taches-cc-resources --skill create-mcp-servers -a codex2Manually: put the files in place
.agents/skills/create-mcp-servers/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add glittercowboy/taches-cc-resources --skill create-mcp-servers -a cursor2Manually: put the files in place
.cursor/skills/create-mcp-servers/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install glittercowboy/taches-cc-resources skills/create-mcp-servers2Manually: put the files in place
.github/skills/create-mcp-servers/SKILL.md1Skills CLI (Vercel)
npx skills add glittercowboy/taches-cc-resources --skill create-mcp-servers -a gemini-cli2Platform CLI
gemini skills install https://github.com/glittercowboy/taches-cc-resources --path skills/create-mcp-serversManually: put the files in place
.devin/skills/create-mcp-servers/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about create-mcp-servers
How do I install create-mcp-servers in Claude Code?
Run both marketplace commands in Claude Code: first /plugin marketplace add glittercowboy/taches-cc-resources, then /plugin install taches-cc-resources@taches-cc-resources.
What does create-mcp-servers do?
It guides Claude Code through an eight-step pipeline to build, test, and register Model Context Protocol servers with tools, resources, and prompts.
Why does the skill require uv for Python servers?
The workflow relies on uv to manage isolated virtual environments and run module commands without dependency conflicts.
Source text
SKILL.md · MIT
---
name: create-mcp-servers
description: Create Model Context Protocol (MCP) servers that expose tools, resources, and prompts to Claude. Use when building custom integrations, APIs, data sources, or any server that Claude should interact with via the MCP protocol. Supports both TypeScript and Python implementations.
---
<objective>
MCP servers extend Claude's capabilities by exposing tools, resources, and prompts. This skill guides creation of production-ready MCP servers with API integrations, OAuth authentication, response optimization, and proper installation in Claude Code and Claude Desktop.
</objective>
<essential_principles>
<the_5_rules>
Every MCP server must follow these:
1. **Never Hardcode Secrets** - Use `${VAR}` expansion in configs, environment variables in code
2. **Use `cwd` Property** - Isolates dependencies (not `--cwd` in args)
3. **Always Absolute Paths** - `which uv` to find paths, never relative
4. **One Server Per Directory** - `~/Developer/mcp/{server-name}/`
5. **Use `uv` for Python** - Better than pip, handles venvs automatically
</the_5_rules>
<security_checklist>
- Never ask user to paste secrets into chat
- Always use environment variables for credentials
- Use ${VAR} expansion in configs
- Provide exact commands for user to run in terminal
- Verify environment variable existence without showing values
- Never hardcode API keys in code or configs
</security_checklist>
<architecture_decision>
Operation count determines architecture:
- **1-2 operations** → Traditional pattern (flat tools)
- **3+ operations** → On-demand discovery pattern (meta-tools)
Traditional: Each operation is a separate tool
On-demand: 4 meta-tools (discover, get_schema, execute, continue) + operations.json
</architecture_decision>
<context>
MCP servers expose:
- **Tools**: Functions Claude can call (API requests, file operations, calculations)
- **Resources**: Data Claude can read (files, database records, API responses)
- **Prompts**: Reusable prompt templates with arguments
Standard location: `~/Developer/mcp/{server-name}/`
</context>
</essential_principles>
<routing>
Based on user intent, route to appropriate workflow:
**No context provided** (skill invoked without description):
Use AskUserQuestion:
- header: "Mode"
- question: "What would you like to do?"
- options:
- "Create a new MCP server" → workflows/create-new-server.md
- "Update an existing MCP server" → workflows/update-existing-server.md
- "Troubleshoot a server" → workflows/troubleshoot-server.md
**Context provided** (user described what they want):
Route directly to workflows/create-new-server.md
</routing>
<workflows_index>
| Workflow | Purpose |
|----------|---------|
| create-new-server.md | Full 8-step workflow from intake to verification |
| update-existing-server.md | Modify or extend an existing server |
| troubleshoot-server.md | Diagnose and fix connection/runtime issues |
</workflows_index>
<templates_index>
| Template | Purpose |
|----------|---------|
| python-server.py | Traditional pattern starter for Python |
| typescript-server.ts | Traditional pattern starter for TypeScript |
| operations.json | On-demand discovery operations definition |
</templates_index>
<scripts_index>
| Script | Purpose |
|--------|---------|
| setup-python-project.sh | Initialize Python MCP project with uv |
| setup-typescript-project.sh | Initialize TypeScript MCP project with npm |
</scripts_index>
<references_index>
**Core workflow:**
- creation-workflow.md - Complete step-by-step with exact commands
**Architecture patterns:**
- traditional-pattern.md - For 1-2 operations (flat tools)
- large-api-pattern.md - For 3+ operations (on-demand discovery)
**Language-specific:**
- python-implementation.md - Async patterns, type hints
- typescript-implementation.md - Type safety, SDK features
**Advanced topics:**
- oauth-implementation.md - OAuth with stdio isolation
- response-optimization.md - Field truncation, pagination
- tools-and-resources.md - Resources API, prompts, streaming
- testing-and-deployment.md - Unit tests, packaging, publishing
- validation-checkpoints.md - All validation checks
- adaptive-questioning-guide.md - Question templates for intake
- api-research-template.md - API research document format
</references_index>
<quick_reference>
```bash
# List servers
claude mcp list
# Add server (Python)
claude mcp add --transport stdio <name> \
--env API_KEY='${API_KEY}' \
-- uv --directory ~/Developer/mcp/<name> run python -m src.server
# Add server (TypeScript)
claude mcp add --transport stdio <name> \
--env API_KEY='${API_KEY}' \
-- node ~/Developer/mcp/<name>/build/index.js
# Remove server
claude mcp remove <name>
# Check logs
tail -f ~/Library/Logs/Claude/mcp-server-<name>.log
# Find paths
which uv && which node && which python
```
</quick_reference>
<troubleshooting_quick>
**Server not appearing:** Check `claude mcp list`, verify config in `~/.claude/settings.json`
**"command not found":** Use absolute paths from `which uv` / `which node`
**Environment variable not found:**
```bash
echo $MY_API_KEY # Check if set
echo 'export MY_API_KEY="value"' >> ~/.zshrc && source ~/.zshrc
```
**Secrets visible in conversation:** STOP. Delete conversation. Rotate credentials. Never paste secrets in chat.
Full troubleshooting: workflows/troubleshoot-server.md
</troubleshooting_quick>
<success_criteria>
A production-ready MCP server has:
- Valid configuration in Claude Code (`claude mcp list` shows ✓ Connected)
- Valid configuration in Claude Desktop config
- Environment variables set securely in ~/.zshrc
- Architecture matches operation count
- OAuth stdio isolation if applicable
- Response optimization for list/search operations
- All validation checkpoints passed
- No errors in logs
</success_criteria>
© Lex Christopherson · MIT · glittercowboy/taches-cc-resources