coderunner
starsforks
How to install coderunner?
instavm/coderunner
$git clone https://github.com/instavm/coderunnerCodeRunner helps you sandbox your AI agents and its actions inside a sandbox.
README.md
| 1 | <div align="center"> |
| 2 | |
| 3 | [](https://github.com/instavm/coderunner/stargazers) |
| 4 | [](https://github.com/instavm/coderunner/blob/master/LICENSE) |
| 5 | </div> |
| 6 | |
| 7 | # CodeRunner: A local sandbox for your AI agents |
| 8 | |
| 9 | CodeRunner helps you sandbox your AI agents and its actions inside a sandbox. |
| 10 | |
| 11 | **Key use case:** You can run multiple Claude Code or AI agents in our sandbox without any fear of data loss and exfilteration. |
| 12 | |
| 13 | |
| 14 | |
| 15 | ## Quick Start |
| 16 | |
| 17 | **Prerequisites:** Mac with macOS and Apple Silicon (M1/M2/M3/M4), Python 3.10+ |
| 18 | |
| 19 | ```bash |
| 20 | git clone https://github.com/instavm/coderunner.git |
| 21 | cd coderunner |
| 22 | chmod +x install.sh |
| 23 | ./install.sh |
| 24 | ``` |
| 25 | |
| 26 | ### Stop and resume |
| 27 | |
| 28 | Stop the sandbox when you are done: |
| 29 | |
| 30 | ```bash |
| 31 | container stop coderunner |
| 32 | ``` |
| 33 | |
| 34 | Resume the same sandbox later, preserving uploads, kernels, and installed packages: |
| 35 | |
| 36 | ```bash |
| 37 | container start coderunner |
| 38 | ``` |
| 39 | |
| 40 | To start over with a clean sandbox, delete the container and run the installer again: |
| 41 | |
| 42 | ```bash |
| 43 | container delete coderunner && ./install.sh |
| 44 | ``` |
| 45 | |
| 46 | |
| 47 | ## Run Claude Code inside a Sandbox |
| 48 | |
| 49 | `./install.sh` (if not already done) |
| 50 | |
| 51 | `container exec -it coderunner /bin/bash` |
| 52 | |
| 53 | `root@coderunner:/app# npm install -g @anthropic-ai/claude-code` |
| 54 | |
| 55 | <img width="741" height="410" alt="image" src="https://github.com/user-attachments/assets/620490cb-4e85-4c37-bc57-ab2fa1762c78" /> |
| 56 | |
| 57 | ## Other Integration Options |
| 58 | |
| 59 | MCP server will be available at: `http://coderunner.local:8222/mcp` |
| 60 | |
| 61 | **Install required packages** (use virtualenv and note the python path): |
| 62 | ```bash |
| 63 | pip install -r examples/requirements.txt |
| 64 | ``` |
| 65 | |
| 66 | #### 1. Claude Desktop Integration |
| 67 | |
| 68 | |
| 69 | <details> |
| 70 | <summary>Configure Claude Desktop to use CodeRunner as an MCP server:</summary> |
| 71 | |
| 72 |  |
| 73 | |
| 74 |  |
| 75 | |
| 76 |  |
| 77 | |
| 78 | 1. **Copy the example configuration:** |
| 79 | ```bash |
| 80 | cd examples |
| 81 | cp claude_desktop/claude_desktop_config.example.json claude_desktop/claude_desktop_config.json |
| 82 | ``` |
| 83 | |
| 84 | 2. **Edit the configuration file** and replace the placeholder paths: |
| 85 | - Replace `/path/to/your/python` with your actual Python path (e.g., `/usr/bin/python3` or `/opt/homebrew/bin/python3`) |
| 86 | - Replace `/path/to/coderunner` with the actual path to your cloned repository |
| 87 | |
| 88 | Example after editing: |
| 89 | ```json |
| 90 | { |
| 91 | "mcpServers": { |
| 92 | "coderunner": { |
| 93 | "command": "/opt/homebrew/bin/python3", |
| 94 | "args": ["/Users/yourname/coderunner/examples/claude_desktop/mcpproxy.py"] |
| 95 | } |
| 96 | } |
| 97 | } |
| 98 | ``` |
| 99 | |
| 100 | 3. **Update Claude Desktop configuration:** |
| 101 | - Open Claude Desktop |
| 102 | - Go to Settings → Developer |
| 103 | - Add the MCP server configuration |
| 104 | - Restart Claude Desktop |
| 105 | |
| 106 | 4. **Start using CodeRunner in Claude:** |
| 107 | You can now ask Claude to execute code, and it will run safely in the sandbox! |
| 108 | </details> |
| 109 | |
| 110 | #### 2. Claude Code CLI |
| 111 | |
| 112 | <details> |
| 113 | <summary>Use CodeRunner with Claude Code CLI for terminal-based AI assistance:</summary> |
| 114 | |
| 115 | **Quick Start:** |
| 116 | |
| 117 | ```bash |
| 118 | # 1. Install and start CodeRunner (one-time setup) |
| 119 | git clone https://github.com/instavm/coderunner.git |
| 120 | cd coderunner |
| 121 | sudo ./install.sh |
| 122 | |
| 123 | # 2. Install the Claude Code plugin |
| 124 | claude plugin marketplace add https://github.com/instavm/coderunner-plugin |
| 125 | claude plugin install instavm-coderunner |
| 126 | |
| 127 | # 3. Reconnect to MCP servers |
| 128 | /mcp |
| 129 | ``` |
| 130 | |
| 131 | **Installation Steps:** |
| 132 | |
| 133 | 1. Navigate to Plugin Marketplace: |
| 134 | |
| 135 |  |
| 136 | |
| 137 | 2. Add the InstaVM repository: |
| 138 | |
| 139 |  |
| 140 | |
| 141 | 3. Execute Python code with Claude Code: |
| 142 | |
| 143 |  |
| 144 | |
| 145 | That's it! Claude Code now has access to all CodeRunner tools: |
| 146 | - **execute_python_code** - Run Python code in persistent Jupyter kernel |
| 147 | - **navigate_and_get_all_visible_text** - Web scraping with Playwright |
| 148 | - **list_skills** - List available skills (docx, xlsx, pptx, pdf, image processing, etc.) |
| 149 | - **get_ |
People who liked this also viewed
6 picksType
- excel-mcpA Model Context Protocol (MCP) server for working with Excel (.xlsx) files without requiring Microsoft Excel. Enables AI assistants to create, read, modify, and format spreadsheets using openpyxl, fully in Python and cross-platform.MCP ServersOct 2025
- memories-with-lessons-mcp-serverI enhance the existing memory mcp server from the official mcp github, so big thanks and credits for creating thisMCP ServersFeb 2025
- anytype-mcpAn MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.MCP ServersJul 2026
- mcp-server-kubernetesMCP Server for kubernetes management commandsMCP ServersJul 2026
- thoughtboxThoughtbox is an intention ledger for agents. Evaluate AI's decisions against its decision-making.MCP ServersJul 2026
- google-drive-mcpA Model Context Protocol (MCP) server that provides secure integration with Google Drive, Docs, Sheets, Slides and Calendar. It allows Claude Desktop and other MCP clients to manage files in Google Drive through a standardized interface.MCP ServersJul 2026
