Host Terminal MCP
Enables AI assistants to execute terminal commands on a host machine with configurable, granular permission controls and safety protections. It features multiple security modes, including allowlists and manual approval, to ensure safe command execution within specified directories.
README
Host Terminal MCP
Run terminal commands on your computer through Claude.
Two transports:
- Stdio — for Claude Desktop and MCP Inspector (default)
- HTTP — for external services calling in over the network (e.g. a chatbot running in Docker)
Quick Start
MCP Server (Claude Desktop / Cowork)
# Install
uv tool install host-terminal-mcp
# Add to Claude Desktop config and restart
Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json on Mac):
{
"mcpServers": {
"host-terminal": {
"command": "host-terminal-mcp"
}
}
}
HTTP Server (External Services)
For services running in Docker or on other machines that need to execute host commands via REST API:
# Install with HTTP extras
uv tool install 'host-terminal-mcp[http]'
# Start in background
nohup host-terminal-mcp --http --port 8099 > /tmp/host-terminal-mcp.log 2>&1 &
# With a specific permission mode
nohup host-terminal-mcp --http --port 8099 --mode ask > /tmp/host-terminal-mcp.log 2>&1 &
# Verify
curl http://localhost:8099/health
Services call POST http://localhost:8099/execute with a JSON body.
From Source
git clone https://github.com/ankitag-in/host-terminal-mcp.git
cd host-terminal-mcp
make install
# MCP stdio server (foreground, for Claude Desktop)
make run
# HTTP server (background daemon)
make start # port 8099, allowlist mode
make start HTTP_PORT=9000 # custom port
make start MODE=ask # ask permission mode
make stop # stop
make status # status + recent logs
make restart # stop + start
HTTP Endpoints
| Endpoint | Method | Purpose |
|---|---|---|
/health |
GET | Health check |
/execute |
POST | Run a command |
/cd |
POST | Change working directory |
/cwd |
GET | Get current directory |
/permissions |
GET | Get permission config |
Example
# Health check
curl http://localhost:8099/health
# Execute a command
curl -X POST http://localhost:8099/execute \
-H "Content-Type: application/json" \
-d '{"command": "docker compose ps", "working_directory": "/path/to/project"}'
# Response
{
"status": "success",
"stdout": "NAME IMAGE ...",
"stderr": "",
"exit_code": 0,
"return_code": 0,
"timed_out": false,
"truncated": false,
"working_directory": "/path/to/project"
}
Use
Ask Claude:
- "List files in my home directory"
- "Show git status"
- "What's running on port 3000?"
Permission Modes
| Mode | Description | Safety |
|---|---|---|
allowlist |
Only pre-approved commands run (default) | Recommended |
ask |
Prompts for unknown commands, can be approved per-session | Use with caution |
allow_all |
Allows everything except blocked commands | Dangerous |
Permission check order: blocked (always wins) > allowed > session-approved > mode decision
Default Allowed Commands
These commands (and their arguments) are allowed out of the box:
File listing & navigation:
ls, ll, la, pwd, tree, find, locate, which, whereis, file
File viewing:
cat, head, tail, less, more, bat, wc
Search:
grep, rg, ag, ack, fzf
Git (read-only):
git status, git log, git diff, git show, git branch, git remote, git tag, git stash list, git rev-parse, git config --get, git config --list, git blame, git shortlog, git describe
System info:
uname, hostname, whoami, id, date, uptime, df, du, free, top -l 1, ps, env, printenv, echo $
Network (read-only):
ping -c, curl -I, curl --head, dig, nslookup, host, ifconfig, ip addr, netstat, ss
Package managers (info only):
npm list, npm ls, npm view, npm show, npm outdated, pip list, pip show, pip freeze, brew list, brew info, apt list, dpkg -l
Dev tool versions:
python --version, python3 --version, node --version, npm --version, cargo --version, rustc --version, go version, java --version, javac --version, ruby --version, docker --version
Docker (read-only):
docker ps, docker images, docker logs
Data processing:
jq, yq
Misc:
man, help, type, stat, md5sum, sha256sum, shasum
Always Blocked Commands
These are blocked regardless of permission mode:
| Pattern | Reason |
|---|---|
rm -rf / |
Recursive delete root |
rm -rf ~ |
Recursive delete home |
rm -rf * |
Recursive delete all |
mkfs |
Format filesystem |
dd if= |
Disk destroyer |
:(){ |
Fork bomb |
> /dev/sd |
Overwrite disk |
chmod -R 777 / |
Dangerous permissions |
chown -R |
Recursive ownership change |
sudo, su, doas |
Privilege escalation |
nc -l |
Netcat listener |
nmap |
Port scanner |
cat /etc/shadow |
Password file |
cat /etc/passwd |
User file |
history -c |
Clear history |
shred |
Secure delete |
Config
Config file: ~/.config/host-terminal-mcp/config.yaml
# Generate default config
host-terminal-mcp --init-config
Add custom allowed commands:
allowed_commands:
# Append your own patterns to the defaults
- pattern: "docker compose logs"
description: "Docker Compose service logs"
- pattern: "docker compose ps"
description: "Docker Compose service status"
- pattern: "docker stats --no-stream"
description: "Docker container resource usage"
- pattern: "redis-cli"
description: "Redis CLI commands"
# Use regex for more flexible matching
- pattern: "^kubectl get "
description: "Kubernetes get resources"
is_regex: true
Other config options:
permission_mode: allowlist # allowlist | ask | allow_all
timeout_seconds: 300 # Max command execution time
max_output_size: 100000 # Max output chars (truncated beyond this)
shell: /bin/bash # Shell to use
allowed_directories: # Commands restricted to these dirs
- /Users/me
environment_passthrough: # Env vars passed to commands
- PATH
- HOME
- USER
- LANG
- LC_ALL
Development
make install # Install all deps (venv auto-created)
make test # Run tests
make lint # Run linters
make format # Format code
make help # Show all targets
License
Apache-2.0
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。