mcp-integration（MCP 集成）

MCP Integration for Claude Code Plugins

Overview

Model Context Protocol (MCP) enables Claude Code plugins to integrate with external services and APIs by providing structured tool access. Use MCP integration to expose external service capabilities as tools within Claude Code.

Key capabilities:

• Connect to external services (databases, APIs, file systems)
• Provide 10+ related tools from a single service
• Handle OAuth and complex authentication flows
• Bundle MCP servers with plugins for automatic setup

MCP Server Configuration Methods

Plugins can bundle MCP servers in two ways:

Method 1: Dedicated.mcp.json (Recommended)

Create .mcp.json at plugin root:

{
  "database-tools": {
    "command": "${CLAUDE_PLUGIN_ROOT}/servers/db-server",
    "args": ["--config", "${CLAUDE_PLUGIN_ROOT}/config.json"],
    "env": {
      "DB_URL": "${DB_URL}"
    }
  }
}

Benefits:

• Clear separation of concerns
• Easier to maintain
• Better for multiple servers

Method 2: Inline in plugin.json

Add mcpServers field to plugin.json:

{
  "name": "my-plugin",
  "version": "1.0.0",
  "mcpServers": {
    "plugin-api": {
      "command": "${CLAUDE_PLUGIN_ROOT}/servers/api-server",
      "args": ["--port", "8080"]
    }
  }
}

Benefits:

• Single configuration file
• Good for simple single-server plugins

MCP Scope System

MCP server configurations follow scope precedence: Local > Project > User.

Plugin-bundled MCP servers (via .mcp.json or inline in plugin.json) auto-start when the plugin is enabled. They interact with user/project MCP configs — if a user has a server with the same name, scope precedence determines which loads.

Discovering MCP Servers

Find existing MCP servers for your plugin using PulseMCP, the comprehensive MCP server directory with 6,800+ servers.

Discovery workflow:

1. Search PulseMCP using Tavily extract on https://www.pulsemcp.com/servers?q=[keyword]
2. Evaluate results by classification (official vs community), popularity, and relevance
3. Fetch detail pages for GitHub links and configuration examples
4. Generate .mcp.json configuration based on server type

See references/server-discovery.md for detailed search instructions, URL patterns, and curated server recommendations by category.

MCP Server Types

stdio (Local Process)

Execute local MCP servers as child processes. Best for local tools and custom servers.

Configuration:

{
  "filesystem": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-filesystem", "/allowed/path"],
    "env": {
      "LOG_LEVEL": "debug"
    }
  }
}

Use cases:

• File system access
• Local database connections
• Custom MCP servers
• NPM-packaged MCP servers

Process management:

• Claude Code spawns and manages the process
• Communicates via stdin/stdout
• Terminates when Claude Code exits

SSE (Server-Sent Events)

Connect to hosted MCP servers with OAuth support. Best for cloud services.

Configuration:

{
  "asana": {
    "type": "sse",
    "url": "https://mcp.asana.com/sse"
  }
}

Use cases:

• Official hosted MCP servers (Asana, GitHub, etc.)
• Cloud services with MCP endpoints
• OAuth-based authentication
• No local installation needed

Authentication:

• OAuth flows handled automatically
• User prompted on first use
• Tokens managed by Claude Code

HTTP (REST API)

Connect to RESTful MCP servers with token authentication.

Configuration:

{
  "api-service": {
    "type": "http",
    "url": "https://api.example.com/mcp",
    "headers": {
      "Authorization": "Bearer ${API_TOKEN}",
      "X-Custom-Header": "value"
    }
  }
}

Use cases:

• REST API-based MCP servers
• Token-based authentication
• Custom API backends
• Stateless interactions

WebSocket (Real-time)

Connect to WebSocket MCP servers for real-time bidirectional communication.

Configuration:

{
  "realtime-service": {
    "type": "ws",
    "url": "wss://mcp.example.com/ws",
    "headers": {
      "Authorization": "Bearer ${TOKEN}"
    }
  }
}

Use cases:

• Real-time data streaming
• Persistent connections
• Push notifications from server
• Low-latency requirements

Environment Variable Expansion

All MCP configurations support environment variable substitution:

${CLAUDE_PLUGIN_ROOT} - Plugin directory (always use for portability):

{
  "command": "${CLAUDE_PLUGIN_ROOT}/servers/my-server"
}

User environment variables - From user's shell:

{
  "env": {
    "API_KEY": "${MY_API_KEY}",
    "DATABASE_URL": "${DB_URL}"
  }
}

Env vars support fallback values: ${VAR:-default_value}. If VAR is unset, default_value is used. Supported in command, args, env, url, and headers fields.

Best practice: Document all required environment variables in plugin README.

MCP Tool Naming

When MCP servers provide tools, they're automatically prefixed:

Format: mcp__plugin_<plugin-name>_<server-name>__<tool-name>

Example:

• Plugin: asana
• Server: asana
• Tool: create_task
• Full name: mcp__plugin_asana_asana__asana_create_task

MCP Resources

MCP servers can expose resources that Claude can access using the @ syntax:

Resource Syntax

@server-name:protocol://path

Examples:

@filesystem:file:///Users/me/project/README.md
@database:postgres://localhost/mydb/users
@github:https://github.com/user/repo

Using Resources in Prompts

Reference resources directly in your prompts:

Look at @filesystem:file:///path/to/config.json and suggest improvements

Claude will fetch the resource content and include it in context.

Resource Types

• file:// - Local file system paths
• https:// - HTTP resources
• Custom protocols - Server-specific (postgres://, s3://, etc.)

MCP Prompts as Commands

MCP servers can expose prompts that appear as slash commands in Claude Code:

Format: /mcp__servername__promptname

Example:

• Server github exposes prompt create-pr
• Available as: /mcp__github__create-pr

MCP prompts appear alongside regular commands in the / menu. They accept arguments and execute the server's prompt template with Claude. This enables MCP servers to provide guided workflows beyond simple tool calls.

Plugin design note: If your MCP server exposes prompts, document their names and expected arguments in your plugin README so users can discover them.

Plugin-provided MCP prompts: If your plugin bundles an MCP server, that server can expose prompts that automatically become slash commands for users. This provides guided workflows beyond simple tool calls — for example, a /mcp__myserver__setup-project prompt that walks users through project configuration.

Tool Search

For MCP servers with many tools, use Tool Search to find relevant tools:

When to use:

• Server provides 10+ tools
• You don't know exact tool names
• Exploring server capabilities

How it works:

1. Claude Code indexes MCP tool names and descriptions
2. Search by natural language or partial names
3. Get filtered list of matching tools

Auto-Enable Behavior

Tool Search activates automatically when MCP servers collectively provide more tools than fit efficiently in Claude's context window (default threshold: 10% of context). Instead of loading all tool descriptions upfront, Claude searches for relevant tools on-demand.

Plugin design implications:

• Many-tool servers: Tools may not be immediately visible; use descriptive tool names and descriptions
• Documentation: Note tool search behavior in README if your server provides 20+ tools
• Environment control: ENABLE_TOOL_SEARCH=auto:5 (custom 5% threshold) or ENABLE_TOOL_SEARCH=false (disable)

This feature is automatic - just ask Claude about available tools or describe what you want to do.

Using MCP Tools in Commands

Pre-allow specific MCP tools in command frontmatter:

---
allowed-tools: mcp__plugin_asana_asana__asana_create_task, mcp__plugin_asana_asana__asana_search_tasks
---

Wildcard (use sparingly):

---
allowed-tools: mcp__plugin_asana_asana__*
---

Best practice: Pre-allow specific tools, not wildcards, for security.

Lifecycle Management

Automatic startup:

• MCP servers start when plugin enables
• Connection established before first tool use
• Restart required for configuration changes

Lifecycle:

1. Plugin loads
2. MCP configuration parsed
3. Server process started (stdio) or connection established (SSE/HTTP/WS)
4. Tools discovered and registered
5. Tools available as mcp__plugin_...__...

Viewing servers: Use /mcp command to see all servers including plugin-provided ones.

Authentication Patterns

OAuth (SSE/HTTP)

OAuth handled automatically by Claude Code:

{
  "type": "sse",
  "url": "https://mcp.example.com/sse"
}

User authenticates in browser on first use. No additional configuration needed.

Token-Based (Headers)

Static or environment variable tokens:

{
  "type": "http",
  "url": "https://api.example.com",
  "headers": {
    "Authorization": "Bearer ${API_TOKEN}"
  }
}

Document required environment variables in README.

Environment Variables (stdio)

Pass configuration to MCP server:

{
  "command": "python",
  "args": ["-m", "my_mcp_server"],
  "env": {
    "DATABASE_URL": "${DB_URL}",
    "API_KEY": "${API_KEY}",
    "LOG_LEVEL": "info"
  }
}

Integration Patterns

Pattern 1: Simple Tool Wrapper

Commands use MCP tools with user interaction:

# Command: create-item.md

---

allowed-tools: `mcp__plugin_name_server__create_item`

Steps:

1. Gather item details from user
2. Use `mcp__plugin_name_server__create_item`
3. Confirm creation

Use for: Adding validation or preprocessing before MCP calls.

Pattern 2: Autonomous Agent

Agents use MCP tools autonomously:

# Agent: data-analyzer.md

Analysis Process:

1. Query data via `mcp__plugin_db_server__query`
2. Process and analyze results
3. Generate insights report

Use for: Multi-step MCP workflows without user interaction.

Pattern 3: Multi-Server Plugin

Integrate multiple MCP servers:

{
  "github": {
    "type": "sse",
    "url": "https://mcp.github.com/sse"
  },
  "jira": {
    "type": "sse",
    "url": "https://mcp.jira.com/sse"
  }
}

Use for: Workflows spanning multiple services.

Security Best Practices

Use HTTPS/WSS

Always use secure connections:

✅ "url": "https://mcp.example.com/sse"
❌ "url": "http://mcp.example.com/sse"

Token Management

DO:

• ✅ Use environment variables for tokens
• ✅ Document required env vars in README
• ✅ Let OAuth flow handle authentication

DON'T:

• ❌ Hardcode tokens in configuration
• ❌ Commit tokens to git
• ❌ Share tokens in documentation

Permission Scoping

Pre-allow only necessary MCP tools:

✅ allowed-tools: `mcp__plugin_api_server__read_data`, `mcp__plugin_api_server__create_item`

❌ allowed-tools: mcp__plugin_api_server__*

Managed MCP Controls (Enterprise)

Organizations can control MCP server access through managed settings.

Place managed-mcp.json at the system-wide managed settings path for exclusive control over MCP server configuration. Alternatively, use allow/deny lists in managed settings:

{
  "allowedMcpServers": [
    { "serverName": "github" },
    { "serverCommand": ["npx", "-y", "@company/mcp-server"] },
    { "serverUrl": "https://mcp.company.com/*" }
  ],
  "deniedMcpServers": [
    { "serverName": "untrusted-server" }
  ]
}

Matcher types:

• serverName — Match by configured server name
• serverCommand — Match by exact command array
• serverUrl — Match by URL pattern (supports * wildcards)

These settings are configured by administrators and cannot be overridden by users or plugins.

Claude Code as MCP Server

Claude Code can itself act as an MCP server, exposing its capabilities to other tools:

claude mcp serve

This enables other MCP-compatible clients to use Claude Code's tools. Useful for building tool chains where Claude Code is one component.

Importing from Claude Desktop

Users who already have MCP servers configured in Claude Desktop can import them:

claude mcp add-from-claude-desktop

This copies MCP server configurations from Claude Desktop into Claude Code. Plugin developers should note that users may already have servers configured this way — avoid name conflicts with commonly used server names.

Dynamic Tool Updates

MCP servers can notify Claude Code when their available tools change at runtime using the list_changed notification. This enables servers that dynamically add or remove tools based on context (e.g., loading project-specific tools after initialization). Claude Code automatically re-discovers tools when list_changed fires, without requiring a restart.

Plugin design note: If your MCP server's available tools depend on runtime state, implement list_changed to ensure Claude Code always has an up-to-date tool list.

MCP Output Limits

MCP tool responses are subject to size limits:

• Warning threshold: 10,000 tokens
• Default maximum: 25,000 tokens (responses exceeding this are truncated)
• Configuration: Set MAX_MCP_OUTPUT_TOKENS environment variable to adjust the maximum

Design MCP tools to return concise, relevant data. Use pagination or filtering for large datasets.

Error Handling

Connection Failures

Handle MCP server unavailability:

• Provide fallback behavior in commands
• Inform user of connection issues
• Check server URL and configuration

Tool Call Errors

Handle failed MCP operations:

• Validate inputs before calling MCP tools
• Provide clear error messages
• Check rate limiting and quotas

Configuration Errors

Validate MCP configuration:

• Test server connectivity during development
• Validate JSON syntax
• Check required environment variables

Performance Considerations

Lazy Loading

MCP servers connect on-demand:

• Not all servers connect at startup
• First tool use triggers connection
• Connection pooling managed automatically

Batching

Batch similar requests when possible:

# Good: Single query with filters
tasks = search_tasks(project="X", assignee="me", limit=50)

# Avoid: Many individual queries
for id in task_ids:
    task = get_task(id)

Testing MCP Integration

Local Testing

1. Configure MCP server in .mcp.json
2. Install plugin locally (.claude-plugin/)
3. Run /mcp to verify server appears
4. Test tool calls in commands
5. Check claude --debug logs for connection issues

Validation Checklist

• MCP configuration is valid JSON
• Server URL is correct and accessible
• Required environment variables documented
• Tools appear in /mcp output
• Authentication works (OAuth or tokens)
• Tool calls succeed from commands
• Error cases handled gracefully

MCP CLI Commands

For testing and managing MCP servers during development:

# Add servers
claude mcp add --transport stdio <name> -- <command> [args...]
claude mcp add --transport http <name> <url>
claude mcp add --transport http <name> <url> --header "Authorization: Bearer token"

# Manage servers
claude mcp list                    # List configured servers
claude mcp get <name>              # Show server details
claude mcp remove <name>           # Remove a server

# Advanced
claude mcp add-json <name> '<json>'           # Add from JSON config
claude mcp add-from-claude-desktop             # Import from Claude Desktop
claude mcp reset-project-choices               # Reset project MCP approval choices

Key flags: --scope (local/project/user), --env KEY=value, --callback-port (for OAuth).

Debugging

Enable Debug Logging

claude --debug

Look for:

• MCP server connection attempts
• Tool discovery logs
• Authentication flows
• Tool call errors

Common Issues

Server not connecting:

• Check URL is correct
• Verify server is running (stdio)
• Check network connectivity
• Review authentication configuration

Tools not available:

• Verify server connected successfully
• Check tool names match exactly
• Run /mcp to see available tools
• Restart Claude Code after config changes

Authentication failing:

• Clear cached auth tokens
• Re-authenticate
• Check token scopes and permissions
• Verify environment variables set

Quick Reference

MCP Server Types

Configuration Checklist

• Server type specified (stdio/SSE/HTTP/ws)
• Type-specific fields complete (command or url)
• Authentication configured
• Environment variables documented
• HTTPS/WSS used (not HTTP/WS)
• ${CLAUDE_PLUGIN_ROOT} used for paths

Best Practices

DO:

• ✅ Use ${CLAUDE_PLUGIN_ROOT} for portable paths
• ✅ Document required environment variables
• ✅ Use secure connections (HTTPS/WSS)
• ✅ Pre-allow specific MCP tools in commands
• ✅ Test MCP integration before publishing
• ✅ Handle connection and tool errors gracefully

DON'T:

• ❌ Hardcode absolute paths
• ❌ Commit credentials to git
• ❌ Use HTTP instead of HTTPS
• ❌ Pre-allow all tools with wildcards
• ❌ Skip error handling
• ❌ Forget to document setup

Additional Resources

Reference Files

For detailed information, consult:

• references/server-discovery.md - Find MCP servers using PulseMCP directory
• references/server-types.md - Deep dive on each server type
• references/authentication.md - Authentication patterns and OAuth
• references/tool-usage.md - Using MCP tools in commands and agents

Example Configurations

Working examples in examples/:

• stdio-server.json - Local stdio MCP server
• sse-server.json - Hosted SSE server with OAuth
• http-server.json - REST API with token auth
• ws-server.json - WebSocket server for real-time communication

External Resources

• Official MCP Docs: https://modelcontextprotocol.io/
• Claude Code MCP Docs: https://docs.claude.com/en/docs/claude-code/mcp
• MCP SDK: @modelcontextprotocol/sdk
• Testing: Use claude --debug and /mcp command

Implementation Workflow

To add MCP integration to a plugin:

1. Choose MCP server type (stdio, SSE, HTTP, ws)
2. Create .mcp.json at plugin root with configuration
3. Use ${CLAUDE_PLUGIN_ROOT} for all file references
4. Document required environment variables in README
5. Test locally with /mcp command
6. Pre-allow MCP tools in relevant commands
7. Handle authentication (OAuth or tokens)
8. Test error cases (connection failures, auth errors)
9. Document MCP integration in plugin README

Focus on stdio for custom/local servers, SSE for hosted services with OAuth.