From claude-code-docs
Claude Code MCP (Model Context Protocol) server configuration reference. Use when adding MCP servers, configuring external tools, or integrating third-party services.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-code-docs:mcp-referenceThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
MCP (Model Context Protocol) connects Claude to external tools and services.
MCP (Model Context Protocol) connects Claude to external tools and services.
| Location | Scope |
|---|---|
~/.claude/settings.json | User (all projects) |
.claude/settings.json | Project (team shared) |
.claude/settings.local.json | Project (gitignored) |
{
"mcpServers": {
"server-name": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-name"],
"env": {
"API_KEY": "your-key"
}
}
}
}
| Field | Required | Description |
|---|---|---|
command | Yes | Executable to run |
args | No | Command arguments |
env | No | Environment variables |
cwd | No | Working directory |
alwaysAllow | No | Array of always-allowed tools |
disabled | No | Boolean to disable server |
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/dir"]
}
}
}
{
"mcpServers": {
"github": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-github"],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_..."
}
}
}
}
{
"mcpServers": {
"postgres": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-postgres"],
"env": {
"DATABASE_URL": "postgresql://..."
}
}
}
}
In plugin's .mcp.json or plugin.json:
{
"mcpServers": {
"plugin-db": {
"command": "${CLAUDE_PLUGIN_ROOT}/server",
"args": ["--config", "${CLAUDE_PLUGIN_ROOT}/config.json"]
}
}
}
Note: Use ${CLAUDE_PLUGIN_ROOT} for plugin paths.
{
"mcpServers": {
"github": {
"command": "...",
"alwaysAllow": [
"create_issue",
"list_repos"
]
}
}
}
If alwaysAllow not specified, Claude asks before using tools.
{
"mcpServers": {
"expensive-server": {
"command": "...",
"disabled": true
}
}
}
{
"env": {
"API_KEY": "${API_KEY}"
}
}
{
"env": {
"DEBUG": "true"
}
}
/mcp
/mcp refresh
Look in /mcp for connection status.
# Debug mode shows server initialization
claude --debug
# Test server manually
npx -y @modelcontextprotocol/server-name
# Check server logs
# Look for MCP-related output in debug
| Issue | Solution |
|---|---|
| Server not starting | Check command exists and is executable |
| Tools not appearing | Verify server implements MCP correctly |
| Permission denied | Check file/directory permissions |
| Environment vars missing | Verify env values are set |
{
"mcpServers": {
"my-python-server": {
"command": "python",
"args": ["-m", "my_mcp_server"],
"cwd": "/path/to/server",
"env": {
"PYTHONPATH": "/path/to/server"
}
}
}
}
.claude/settings.local.json for sensitive configsalwaysAllow to necessary tools onlyFor complete documentation, see:
npx claudepluginhub DuncanJurman/entropy-plugins --plugin claude-code-docsFetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
Applies a firm's KYC/AML rules grid to parsed onboarding records: assigns risk rating, checks required documents, outputs rule outcomes with citations, and routes for escalation.
Generates daily or weekly digests of activity from connected sources (chat, email, docs, tasks, CRM), highlighting action items, decisions, mentions, and project updates.