Enjoy DeepSeek-Powered AI Coding Experience

Deep Code is a terminal AI coding assistant built for DeepSeek-V4 — deep thinking, Agent Skills, and both a CLI and a VS Code extension.

Install Deep Code CLI Install VS Code Extension Join Community Read Docs Deep Code Plus

DeepSeek Priority|Persistent Sessions|Read/Write/Edit/Execute|Agent Skills|OpenAI Compatible

Built for DeepSeek Models

DeepSeek Priority

Ultra-Long Context

Tool Calling

Skill Extension

Model Compatible

Complete Coding Loop in Terminal

Ask, modify, and execute in the same context without switching windows

Persistent Sessions

1M ultra-long context

Tool-Driven Workflow

Read/Write/Edit/Execute commands

DeepSeek Priority

Stable access, no proxy needed

Beyond Chat — Deep into Real Coding Tasks

DeepSeek Optimized

Project-Level Memory

Code Read/Write/Execute

Skill Marketplace

Immersive Interaction

Multi-Model Compatible

Three Steps to Complete Coding Tasks

Three-step flow: connect model, ask in terminal, auto-generate code

Handle All Common Dev Tasks

Code Comprehension
Document Translation
Continuous Task Execution
Custom Model Integration

Deep Code Plus Subscription

Built for individual developers and everyday work: long-term, high-frequency text and code, plus image understanding, image generation, and video generation.

Loading…

Why Choose Deep Code over Regular AI Tools?

Regular AI Tools

Deep Code

Misses DeepSeek features like thinking mode, reasoning control, and prefix caching.

DeepSeek Priority, Ultra-Long Context, File Operations, Skill Extension, Model Compatible

Get Started in Minutes

Create ~/.deepcode/settings.json with:

{
  "env": {
    "MODEL": "deepseek-v4-pro",
    "BASE_URL": "https://api.deepseek.com",
    "API_KEY": "sk-..."
  },
  "thinkingEnabled": true,
  "reasoningEffort": "max"
}

Install Deep Code CLI Install VS Code Extension Join Community

Frequently Asked Questions

Does Deep Code have a VSCode extension?

Yes. Deep Code offers a full-featured VSCode extension, available on the VSCode Marketplace. The extension shares the ~/.deepcode/settings.json configuration file with the CLI, so you can switch seamlessly between the terminal and the editor.

Does Deep Code support understanding images?

Yes. Multimodal models such as deepseek-flash can read local images directly, or you can paste images from the clipboard with Ctrl+V. Other models use the built-in image-understanding tool automatically. See docs/configuration_en.md for configuration details.

How do I get notified (e.g., in Slack) when a task completes?

Write a shell notification script — for example, one that calls a Slack webhook — then set the notify field in ~/.deepcode/settings.json to the full path of the script. Deep Code runs it automatically when a task completes. For detailed steps, see docs/notify_en.md.

How do I enable web search?

Deep Code comes with a built-in, free Web Search tool that works well for most use cases. If you prefer to use a custom script for web search, set the webSearchTool field in ~/.deepcode/settings.json to the full path of your script. For detailed steps, refer to: https://github.com/qorzj/web_search_cli

Does it support Coding Plan?

Yes. Just set env.BASE_URL in ~/.deepcode/settings.json to an OpenAI-compatible API endpoint. Take Volcengine Ark's Coding Plan as an example:

{
  "env": {
    "MODEL": "ark-code-latest",
    "BASE_URL": "<your Coding Plan endpoint>",
    "API_KEY": "**************"
  },
  "thinkingEnabled": true
}

How do I configure MCP?

Deep Code supports MCP (Model Context Protocol) to connect external services such as GitHub, browsers, databases, and more. Configure the mcpServers field in settings.json to enable it, then use the /mcp command to view MCP server status and available tools. For detailed setup instructions, see docs/mcp.md.

Does Deep Code only support YOLO mode?

No. Deep Code has a built-in fine-grained permission control mechanism that lets you confirm operations before the AI assistant executes shell commands, reads/writes files, accesses the network, and more. You can configure each permission scope's policy — always allow, always ask, or deny — via the permissions field in settings.json. See docs/permission.md for details.