Command Line Agentic Code Assistants on Windows

All Require NodeJS:

• Internet connection
• Node.js 20+

winget install OpenJS.NodeJS # WinGet via Windows PowerShell/pwsh

curl -qL https://www.npmjs.com/install.sh | sh # Shell Script via bash/zsh

---

Claude Code (Anthropic)

Installation:

• npm install -g @anthropic-ai/claude-code
• Navigate to project directory
• Run claude
• Complete OAuth authentication (requires subscription or api billing)

Authentication:

• Sign in CLaude Code with Claude Pro/Max Subscription
• Or purchase API at Anthropic Console console.anthropic.com

Structure:

• Config: ~\.claude
• Binaries: ~\AppData\Roaming\npm\node_modules\@anthropic-ai\claude-code
• Operations: ~\AppData\Roaming\Claude
• Prompt instructions: ~\.claude\claude.md (global) or .\claude.md (project-specific)

---

Gemini CLI (Google)

Installation:

• npm install -g @google/gemini-cli
• Navigate to project directory
• Run gemini
• Use a generated API key (requires API billing)

Authentication:

• Generate an API key at ai.google.dev/aistudio
• Add the key to a .env file in your project directory: GEMINI_API_KEY="YOUR_API_KEY"

Structure:

• Config: ~\.gemini
• Binaries: ~\AppData\Roaming\npm\node_modules\@google\gemini-cli
• Operations: ~\AppData\Roaming\Gemini
• Prompt instructions: ~\.gemini\gemini.md (global) or .\gemini.md (project-specific)

---

GROK CLI (xAI)

Installation:

• npm install -g @vibe-kit/grok-cli
• Navigate to project directory
• Run grok
• Use generated API key (requires subscription or api billing)

Authentication:

• Generate API key at console.x.ai/api
• Sign in with Grok/SuperGrok subscription
• Or purcahse API key without login
• Set GROK_API_KEY environment variable, use --api-key flag, or save to ~/.grok/user-settings.json

Structure:

• Config: ~\.grok
• Binaries: ~\AppData\Roaming\npm\node_modules\@vibe-kit\grok-cli
• Operations: ~\AppData\Roaming\GROK
• Prompt instructions: ~\.grok\GROK.md (global) or .\GROK.md (project-specific)

---

OpenCode CLI (SST)

Installation:

• npm install -g opencode-ai@latest (or curl -fsSL https://opencode.ai/install | bash)
• Navigate to project directory
• Run opencode
• Complete authentication as prompted

Authentication:

• Sign up or login at opencode.ai
• Follow the authentication flow when running the CLI for the first time
• API key will be stored in configuration file

Structure:

• Config: ~\.opencode
• Binaries: ~\AppData\Roaming\npm\node_modules\opencode-ai
• Operations: ~\AppData\Roaming\OpenCode
• Prompt instructions: ~\.opencode\opencode.md (global) or .\opencode.md (project-specific)

---

Qwen Code (Alibaba)

Installation:

• Use Sign Up to create an account on chat.qwen.ai
• npm install -g @qwen-code/qwen-code@latest
• Navigate to project directory
• Run qwen

Authentication:

• Login using 1. Qwen OAuth (creates ~\.qwen\oauth_creds.json)
• Select 2. OpenAI to configure OpenAI compatible API LLM source (e.x. Ollama http://localhost:11434/v1)

Structure:

• Config: ~\.qwen
• Binaries: ~\AppData\Roaming\npm\node_modules\@qwen-code\qwen-code
• Operations: ~\AppData\Roaming\qwen
• Prompt instructions: ~\.qwen\qwen.md (global) or .\qwen.md (project-specific)

---

Copilot Codex CLI (via GitHub CLI)

Installation:

• GitHub CLI (winget install GitHub.cli)
• gh extension install github/gh-copilot
• Navigate to project directory
• Authenticate via gh auth login (GitHub token with Copilot access)
• Navigate to project directory
• Run gh copilot to begin interaction

Authentication:

• Requires GitHub Copilot subscription
• Uses GitHub OAuth (gh auth login)
• Verify access at https://github.com/settings/copilot

Structure:

• Config: ~\.config\gh\copilot-cli.json
• Binaries: ~\AppData\Local\GitHubCopilotCLI or ~\AppData\Local\GitHub\CLI\extensions
• Operations: Local interaction via gh copilot [prompt] or interactive shell
• Prompt instructions: Uses inline CLI or external *.md prompt templates (optional)

---

Codex CLI (OpenAI)

Installation:

• Requires Window Subsystem for Linux (WSL2) with NodeL2 with Node.js (v14+) installed
• From your WSL2 terminal, run npm install -g @openai/codex
• Navigate to your project directory: pushd ~/projects/my-codex-app
• Set API key export OPENAI_API_KEY="<YourKeyHere> (see below for details on obtaining an API key)
• Run codex

Authentication:

• Set OpenAI API key export OPENAI_API_KEY="<YourKeyHere>
• Get your API key from https://platform.openai.com/api-keys

Structure:

• Config: ~/.codex/config.toml — includes model settings, timeout, and CLI behavior
• Binaries: ~/.npm-global/bin/codex or /usr/local/bin/codex
• Prompt instructions: .codex/instructions.md (global) or ./CODEX.md (project-specific)