by holasoymalva
DeepSeek CLI, a command-line AI coding assistant that leverages the powerful DeepSeek Coder models
# Add to your Claude Code skills
git clone https://github.com/holasoymalva/deepseek-cliLast scanned: 5/30/2026
{
"issues": [
{
"type": "npm-audit",
"message": "axios: Axios is vulnerable to DoS attack through lack of data size check",
"severity": "high"
},
{
"type": "npm-audit",
"message": "diff: jsdiff has a Denial of Service vulnerability in parsePatch and applyPatch",
"severity": "low"
},
{
"type": "npm-audit",
"message": "follow-redirects: follow-redirects leaks Custom Authentication Headers to Cross-Domain Redirect Targets",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "form-data: form-data uses unsafe random function in form-data for choosing boundary",
"severity": "critical"
}
],
"status": "FAILED",
"scannedAt": "2026-05-30T15:19:58.668Z",
"npmAuditRan": true,
"pipAuditRan": true
}deepseek-cli is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by holasoymalva. DeepSeek CLI, a command-line AI coding assistant that leverages the powerful DeepSeek Coder models. It has 290 GitHub stars.
deepseek-cli failed SkillsLLM's automated security scan, which flagged one or more high-severity issues. Review the Security Report section carefully before using it.
Clone the repository with "git clone https://github.com/holasoymalva/deepseek-cli" and add it to your Claude Code skills directory (see the Installation section above).
deepseek-cli is primarily written in TypeScript. It is open-source under holasoymalva on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh deepseek-cli against similar tools.
No comments yet. Be the first to share your thoughts!
Requires a passing catalog security scan. Resolve the flagged issues and resubmit to enable featuring.
This repository contains the DeepSeek CLI, a command-line AI coding assistant that leverages the powerful DeepSeek Coder models to accelerate your development workflows and enhance your coding experience.
With the DeepSeek CLI you can:
Prerequisites:
Install Ollama:
# macOS
brew install ollama
# Linux
curl -fsSL https://ollama.ai/install.sh | sh
# Windows: Download from https://ollama.ai
Install the CLI:
npm install -g run-deepseek-cli
Setup Local Environment:
# Start Ollama service
ollama serve
# Install DeepSeek Coder model (choose one)
ollama pull deepseek-coder:6.7b # Recommended (4GB)
ollama pull deepseek-coder:1.3b # Lightweight (1GB)
ollama pull deepseek-coder:33b # Most capable (19GB)
Start Using:
deepseek
Install the CLI:
npm install -g run-deepseek-cli
Configure API Access:
export DEEPSEEK_API_KEY="your_api_key_here"
export DEEPSEEK_USE_LOCAL=false
Get your API key from DeepSeek Platform.
Start Using:
deepseek
Once the CLI is running, you can start interacting with DeepSeek Coder from your shell.
cd new-project/
deepseek
> Create a FastAPI web application with user authentication and a PostgreSQL database
git clone https://github.com/your-repo/existing-project
cd existing-project
deepseek
> Analyze this codebase and suggest performance optimizations
deepseek
> Write a Python function to implement binary search with proper error handling
deepseek
> Convert this JavaScript function to TypeScript with proper type annotations
Start by cding into an existing or newly-cloned repository and running deepseek.
> Explain the architecture of this application and identify the main components.
> What design patterns are used in this codebase?
> Find potential security vulnerabilities in this code.
> Implement a REST API for user management with CRUD operations.
> Refactor this class to follow SOLID principles.
> Add comprehensive error handling to this module.
> Help me migrate this React class component to hooks.
> Convert this Python 2.7 code to Python 3.10 with type hints.
> Upgrade this Express.js app to use async/await instead of callbacks.
> Review this pull request and suggest improvements.
> Add unit tests for all functions in this file.
> Optimize this algorithm for better time complexity.
> Create a React TypeScript project with Redux Toolkit and Material-UI.
> Set up a Django project with Docker, PostgreSQL, and Redis.
> Generate a microservices architecture using Node.js and Docker Compose.
| Variable | Description | Default |
|---|---|---|
DEEPSEEK_USE_LOCAL |
Use local Ollama instead of cloud API | true |
DEEPSEEK_MODEL |
Model to use | deepseek-coder:6.7b |
OLLAMA_HOST |
Ollama server URL | http://localhost:11434 |
DEEPSEEK_API_KEY |
DeepSeek API key (cloud mode only) | - |
| Model | Size | Memory | Best For |
|---|---|---|---|
deepseek-coder:1.3b |
1GB | 2GB RAM | Quick completions, lightweight tasks |
deepseek-coder:6.7b |
4GB | 8GB RAM | General coding, recommended |
deepseek-coder:33b |
19GB | 32GB RAM | Complex analysis, best quality |
Create a .env file in your project root:
# Local mode (default)
DEEPSEEK_USE_LOCAL=true
DEEPSEEK_MODEL=deepseek-coder:6.7b
OLLAMA_HOST=http://localhost:11434
# Cloud mode (optional)
# DEEPSEEK_USE_LOCAL=false
# DEEPSEEK_API_KEY=your_api_key_here
DeepSeek CLI supports intelligent code generation and completion for 100+ programming languages including:
Popular Languages:
Python, JavaScript, TypeScript, Java, C++, C#, Go, Rust, PHP, Ruby, Kotlin, Swift, Scala, R, Julia, Dart, HTML, CSS, SQL
Systems & Scripting:
Shell, PowerShell, Bash, Dockerfile, Makefile, YAML, JSON, XML
Specialized:
CUDA, Assembly, Verilog, VHDL, Solidity, Protocol Buffer, Thrift
For the complete list, see Supported Languages.
| Command | Description |
|---|---|
deepseek |
Start interactive mode |
deepseek setup |
Setup local Ollama environment |
deepseek chat "prompt" |
Single prompt mode |
deepseek --local |
Force local mode |
deepseek --model <model> |
Specify model to use |
deepseek --help |
Show help information |
deepseek --version |
Show version |
The setup command helps you install and configure the local environment:
deepseek setup
This will:
The interactive mode provides a rich REPL experience:
DeepSeek CLI automatically analyzes your repository structure and includes relevant context:
# Analyze entire repository
deepseek --include-all
> Suggest improvements for the overall code architecture
# Include specific files/directories
deepseek --include src/ tests/ README.md
> Update the documentation to reflect recent API changes
Generate common patterns and boilerplate:
> Generate a React component template with props validation
> Create a Python class with proper docstrings and type hints
> Set up a GitHub Actions workflow for CI/CD
> Analyze the git diff and write a commit message
> Generate API documentation from this OpenAPI spec
> Create database migration scripts from these model changes
npm install -g run-deepseek-cli
git clone https://github.com/your-username/deepseek-cli.git
cd deepseek-cli
npm install
npm run build
npm link
docker run -it -v $(pwd):/workspace -e DEEPSEEK_API_KEY=your_key deepseek/cli
We welcome contributions! Please see our Contributing Guide for details.
npm installnpm testnpm run devnpm run build
npm run package
| Model | Speed | Context Window | Best For |
|---|---|---|---|
| deepseek-coder-1.3b-instruct | Fastest | 16K tokens | Quick completions |
| deepseek-coder-6.7b-instruct | Fast | 16K tokens | General coding tasks |
| deepseek-coder-33b-instruct | Slower | 16K tokens | Complex analysis |
This project leverages the DeepSeek Coder models:
For more details about the models, visit the DeepSeek Coder GitHub repository.
API Key Issues: