# Container Use Extension
> Use Container-Use MCP as a goose Extension
- 网址：https://funcoding.ai/agents/goose/mcp/container-use-mcp/
- 来源：goose 官方文档原文（英文），Apache-2.0 许可，同步于 2026-10-11
- 官方原文：https://goose-docs.ai/docs/mcp/container-use-mcp
---
This tutorial covers how to add the [Container Use MCP Server](https://container-use.com) as a goose extension enabling goose to work in an isolated environment.
**Quick Install**
**goose Desktop**
[Launch the installer](goose://extension?cmd=container-use&arg=stdio&id=container-use&name=container%20use&description=use%20containers%20with%20dagger%20and%20git%20for%20isolated%20environments)
**goose CLI**
**Command**
```sh
container-use stdio
```
## Configuration
You'll need [Docker](https://www.docker.com/) installed on your system. If you want to use the remote MCP server, you'll also need [Node.js](https://nodejs.org/) installed.
**Remote MCP**
**goose Desktop**
**goose CLI**
**Local MCP**
**goose Desktop**
**goose CLI**
## Example Usage
Here's a simple example of updating a file-based to-do app to use a SQLite database plus full testing - all within an isolated container environment.
See [Isolated Development Environments](https://funcoding.ai/agents/goose/tutorials/isolated-development-environments/) for more examples, common workflows, and troubleshooting tips.
#### Prompt
```
Run a container agent to add a feature to save my to-do list data in sqlite, build and run tests, but use a separate Git branch so my main code stays safe.
```
#### goose Output
```
I'll help you create a container environment to add SQLite functionality to your to-do list application on a separate Git branch. Let me start by creating the environment and examining your current code structure.
─── environment_create | containeruse ──────────────────────────
environment_source: ./examples/todo-list
explanation: ...
title: ...
Now let me set up the environment with the necessary tools and then proceed with the SQLite feature development:
─── environment_config | containeruse ──────────────────────────
config:
base_image: python:3.11-slim
setup_commands:
-
[REDACTED: 63 chars]
-
pip install pytest pytest-cov
environment_id: thankful-whale
environment_source: ./examples/todo-list
explanation: ...
Let me install pytest after the environment is running:
---
Great! I've created a container environment for safely adding SQLite functionality to your to-do list app. Let me first explore the current code structure and then create a new Git branch.
---
Now I'll create the SQLite-based storage implementation. First, let me create a new version of the TaskManager with SQLite support:
---
## ✅ SQLite Feature Successfully Added!
I've successfully created a container environment and added SQLite functionality to your to-do list application! Here's what was accomplished:
### 🎯 **Features Implemented**
1. **SQLite Database Storage**
- Replaced plain text file storage with SQLite database
- Created proper database schema with auto-incrementing IDs
2. **Enhanced Task Management**
- All original functionality preserved (add, list, mark done/undone, delete, clear)
- Added new **statistics feature** showing completion rates
- Improved task display with database IDs
3. **Two Implementation Versions**
- `taskman_sqlite.py` - Full-featured with Rich UI
- `taskman_sqlite_simple.py` - Simplified without external dependencies
### 🧪 **Testing Completed**
- ✅ Created comprehensive test suite (`test_taskman_sqlite.py`)
- ✅ Built simple test runner (`test_simple.py`)
- ✅ All tests pass successfully
- ✅ Manual testing confirms functionality works perfectly
### 📊 **Live Demo Results**
# Added tasks successfully
✓ Added task: Buy groceries
✓ Added task: Walk the dog
# Listed tasks with SQLite IDs
📋 Task List (SQLite)
# ID Status Task
1 1 ✓ Done Buy groceries
2 2 ○ Todo Walk the dog
# Statistics working
Task Statistics:
Total tasks: 2
Completed: 1
Pending: 1
Completion rate: 50.0%
# Database verified
sqlite3 tasks.db "SELECT * FROM tasks;"
1|Buy groceries|1
2|Walk the dog|0
### 🔒 **Safe Development**
- ✅ Used separate container environment
- ✅ All changes isolated from your main code
- ✅ Original implementation preserved
- ✅ Backward compatibility maintained
### 📁 **Files Created**
- `taskman_sqlite_simple.py` - Main SQLite implementation
- `test_simple.py` - Test suite
- `SQLITE_FEATURE_README.md` - Comprehensive documentation
- `tasks.db` - SQLite database file
The SQLite implementation is fully functional and ready for use! The original file-based version remains untouched, so you can safely evaluate the new SQLite features without any risk to your existing code.
```