Local-CLI

Local CLI

npm version GitHub release License: MIT Node.js

OpenAI-Compatible CLI Coding Agent for Local & On-Prem LLM Environments

Use your own LLM (vLLM, Ollama, LM Studio, Azure OpenAI, or any OpenAI-compatible API) as a full coding agent - no cloud dependency, no API key costs.

Quick Demo — Tool Usage

Quick Demo

Quick Demo: Watch Local CLI use tools to automate real tasks.

https://github.com/user-attachments/assets/77cc96c9-cb22-4411-8744-3a006b00c580

Plan & Execute Demo: Local CLI autonomously editing code with Plan & Execute.

Office Automation Demos

Create professional Word, Excel, and PowerPoint documents with AI:

Word Creation Demo
Word Document Creation 		Excel Creation Demo
Excel Spreadsheet Creation
PowerPoint Creation Demo
PowerPoint Presentation Creation 		Windows Auto Update Demo
Windows Auto Update

Why Local CLI?

Benefit Description
Zero Cloud Dependency Runs entirely on your local/on-prem LLM. Your code never leaves your network.
No API Cost Use open-source models (Llama, Qwen, DeepSeek, etc.) for free.
Any OpenAI-Compatible API Works with vLLM, Ollama, LM Studio, Azure OpenAI, Google Gemini, and more.
Autonomous Coding Agent Reads, searches, edits, and creates code files — not just chat.
Plan & Execute Breaks complex tasks into TODO steps and executes them step by step.
Safe by Default Supervised mode requires your approval before any file modification.
Office Sub-Agents Dedicated create/modify sub-agents for Excel, Word, PowerPoint with structured execution.
Pipe Mode Non-interactive CLI mode (-p) for scripting and automation pipelines.
Desktop GUI (Electron) Dual-window desktop app with chat + real-time task monitoring. Auto-update supported.
Vision Model Support Analyze images and screenshots with Vision Language Models.
Office Automation Control Excel, Word, PowerPoint directly via PowerShell/COM (Windows).
Browser Automation Chrome/Edge CDP control - navigate, click, screenshot, scrape data.

Installation

CLI (npm)

# Install globally
npm install -g local-cli-agent

# Run
local-cli

The endpoint setup wizard launches automatically on first run.

Desktop App (Windows)

Download the latest LOCAL-BOT-Setup-{version}.exe from the Releases page.

Build from Source

git clone https://github.com/A2G-Dev-Space/Local-CLI.git
cd Local-CLI
npm install && npm run build
node dist/cli.js

Key Features

Dual-Window Desktop App

The Electron desktop app provides a Chat Window and a separate Task Popup for real-time task monitoring:

Plan & Execute

Automatically decomposes requests into TODO steps and executes them:

You: Add a logging system to the project

TODO List                            1/3
  [x] Create logger.ts
  [ ] Add imports to existing code
  [ ] Apply error handling

Deep Research (Search Sub-Agent)

Performs Perplexity-level web research using its own headless Chrome engine — no external search API (Tavily, SerpAPI, Google API) required.

Confluence Integration

Edit or create Confluence pages directly via confluence_request:

{
  "browserServices": [{ "type": "confluence", "name": "My Confluence", "url": "https://confluence.example.com" }],
  "researchUrls": [{ "name": "My Confluence", "url": "https://confluence.example.com" }]
}

Jira Integration

Manage Jira issues directly via jira_request:

{
  "browserServices": [{ "type": "jira", "name": "My Jira", "url": "https://jira.example.com" }]
}

Office Sub-Agents

Office automation uses a dedicated Sub-Agent architecture where each Office app has specialized create and modify agents:

App Create Agent Modify Agent Capabilities
Excel Structured sheet builder Tool-based editor Charts, formatting, conditional formatting, pivot tables, formulas, sparklines
Word Section-by-section builder Tool-based editor Headers, paragraphs, tables, images, TOC, footnotes, styles, page setup
PowerPoint Layout-aware slide builder Tool-based editor Slides, text, images, shapes, themes, speaker notes, transitions

How it works:

  1. Enhancement LLM analyzes the user’s request and generates detailed content
  2. Planning LLM creates a structured JSON plan (design system, sections/sheets/slides)
  3. Execution builds the document section-by-section using specialized builder functions
  4. Review LLM evaluates quality and triggers refinements if needed

This architecture produces enterprise-quality Office documents with consistent formatting, proper structure, and content appropriate to the topic.

Vision Language Model (VLM)

Analyze images and screenshots directly from the chat:

Supervised Mode

Every file modification requires your explicit approval:

Browser Automation

LLM Compatibility

Works well even with smaller or less capable open-source models:

Session Management

Commands & Shortcuts

Slash Commands

Command Description
/help Show help information
/clear Reset conversation
/compact Compress conversation
/load Load saved session
/model Switch model
/settings Settings menu
/usage Token usage
/docs Docs management
/tool Toggle optional tools (browser/office)

Keyboard Shortcuts

Key Action
Ctrl+C Exit
ESC Interrupt current task
Tab Toggle Auto / Supervised
@ File browser
/ Command autocomplete

Configuration

# Terminal — setup wizard launches on first run
local-cli

# Inside Local CLI — open settings menu
/settings

Any OpenAI-compatible API works: vLLM, Ollama, LM Studio, Azure OpenAI, Google Gemini, or internal LLM servers.

Requirements

Changelog

v5.0.4

v5.0.2

v5.0.1

v5.0.0

v4.5.1

v4.5.0

v4.4.0

v4.3.0

v4.2.0

v4.1.5 — v4.1.7

Documentation

Contact

License

MIT License

GitHub: https://github.com/A2G-Dev-Space/Local-CLI

This site is open source. Improve this page.