Building agentic coding assistant in public

Langcli

Langcli is an interactive AI coding assistant in the terminal, built upon the leaked code of Claude Code. Therefore, the way to use Langcli is exactly the same as that of standard Claude Code. Even more exciting is that Langcli is deeply integrated with LangRouter, allowing you to friendly use and switch between mainstream LLM models (including Claude Opus 4.6, GLM 5.1, Kimi K2.5, Minimax M2.5, Deepseek v3.2, etc.) within an ongoing session as needed, without interrupting your context.

Why not integrate with OpenRouter? OpenRouter's API for DeepSeek models is too expensive! Moreover, OpenRouter's Anthropic API interface is not as compatible with the official Claude Code API interface.

Installation

Quick Install (Recommended)

Linux / macOS

bash -c "$(curl -fsSL https://assets.langcli.com/installation/install-langcli.sh)"

Windows (Run as Administrator CMD)

curl -fsSL -o %TEMP%\install-langcli.bat https://assets.langcli.com/installation/install-langcli.bat && %TEMP%\install-langcli.bat

Note: It's recommended to restart your terminal after installation to ensure environment variables take effect.

Manual Installation

Prerequisites

Make sure you have Node.js 20 or later installed. Download it from nodejs.org.

NPM

npm i -g langcli-com

Quick Start

LangRouter API Key Preparation

Go to LangRouter, register an account, save your API key.

Running

# Start Langcli (interactive)
langcli

# Then, in the session:
/help

Compile from source code yourself (optional)

Environmental requirements

• Bun >= 1.3.11

Install dependencies

bun install

Running

# Development mode
bun run dev

# Build and run
bun run build && bun dist/cli.js

The built version can be started wi