Skip to main content

HKU's CLI-Anything Turns Any Software into AI-Friendly Tools with One Command

HKU Team Revolutionizes AI-Software Integration

Imagine telling an AI to edit a photo in GIMP or render a 3D model in Blender without worrying about the interface crashing. That future just got closer thanks to CLI-Anything, a groundbreaking open-source project from the University of Hong Kong's Data Intelligence Lab.

Image

The Problem with Current Approaches

For developers working with AI agents, integrating professional software has been a persistent headache. Traditional UI automation solutions are notoriously fragile - one minor update can break everything. Meanwhile, writing custom integration scripts consumes valuable time that could be spent on actual innovation.

"We kept hearing the same frustrations from developers," explains the HKU team behind the project. "They wanted to leverage powerful tools like Blender or LibreOffice through AI, but the existing methods were either unreliable or required too much manual work."

How CLI-Anything Works Its Magic

The solution is surprisingly elegant. CLI-Anything automatically analyzes software source code and generates:

  • Structured command-line interfaces that AI can understand and chain together
  • JSON output support for easy parsing by machine learning models
  • Self-documenting commands with standard --help functionality
  • Interactive REPL mode for testing and debugging
  • Rock-solid reliability backed by over 1,500 tests

What makes this different from traditional automation? Instead of simulating mouse clicks and keyboard inputs (the approach used by most RPA tools), CLI-Anything taps directly into the software's core functionality. This means no more worrying about window positions changing or buttons moving in an update.

Growing List of Supported Applications

The team has already adapted 13 major software packages including:

  • Creative tools: GIMP, Blender, Inkscape, Audacity
  • Productivity apps: LibreOffice, Draw.io, Mermaid
  • Video software: OBS Studio, Kdenlive, Shotcut
  • AI utilities: ComfyUI for image generation

The roadmap includes expanding into CAD design tools, digital audio workstations (DAWs), and scientific computing applications. Community contributions for additional software integrations are actively encouraged.

Easy Setup for Developers

The installation process couldn't be simpler:

/plugin marketplace add HKUDS/CLI-Anything
/plugin install cli-anything
/cli-anything 

The tool works seamlessly with popular agent frameworks like OpenClaw and OpenCode while running entirely locally - no cloud dependencies or complicated configurations required.

Why This Matters for AI Development

The enthusiastic response from developers (over 17,000 GitHub stars already) speaks volumes. CLI-Anything isn't just another utility - it's foundational infrastructure for the emerging world where AI agents routinely interact with professional software. Early adopters are already automating complex workflows like batch image processing in GIMP or generating 3D model renders through simple text commands to their AI assistants.

The project represents a significant step toward what many see as inevitable: a future where we increasingly delegate software operation to intelligent agents rather than wrestling with interfaces ourselves. Project Address: https://github.com/HKUDS/CLI-Anything

Enjoyed this article?

Subscribe to our newsletter for the latest AI news, product reviews, and project recommendations delivered to your inbox weekly.

Weekly digestFree foreverUnsubscribe anytime

Related Articles

News

Mistral AI's Small4: A Triple-Threat Open Source Model Arrives

Mistral AI has unveiled its latest open-source marvel - the Small4 model. This isn't just another incremental update; it combines three powerful capabilities into one package: logical reasoning, multimodal processing, and coding assistance. With its efficient 128-expert architecture and configurable performance modes, developers now have a versatile tool that adapts to different needs while cutting computational costs.

March 17, 2026
AI modelsopen sourceMistral AI
News

Baidu's Miaoda Makes App Development Accessible to All

Baidu has unveiled its Miaoda Application Generation Skill, allowing users worldwide to create apps with minimal technical know-how. The platform simplifies development into three straightforward steps, already serving over 10 million users and generating billions in value. Notably, it's empowering solo entrepreneurs to build profitable businesses with AI tools.

March 17, 2026
AI developmentBaiduno-code platforms
NVIDIA's Nemotron 3 Series: AI Gets a Fivefold Speed Boost
News

NVIDIA's Nemotron 3 Series: AI Gets a Fivefold Speed Boost

At the 2026 GTC conference, NVIDIA unveiled its Nemotron 3 series of open-source AI models, with the flagship Ultra version delivering five times faster processing. The release also includes innovative multimodal tools for audio-visual integration and real-time conversation, plus breakthroughs in robotics and medical research. Major industry players are already adopting these cutting-edge technologies.

March 17, 2026
AI innovationNVIDIAmachine learning
Tsinghua's AI Classroom Brings Learning to Life
News

Tsinghua's AI Classroom Brings Learning to Life

Tsinghua University has unveiled OpenMAIC, an innovative open-source platform that transforms any topic into a dynamic virtual classroom. Unlike traditional AI tutors, this system creates a complete learning ecosystem with multiple AI roles - from teachers to classmates - making education more interactive and engaging. Already tested with 500 students, the technology promises to democratize quality education globally.

March 16, 2026
AI educationvirtual classroomopen source
News

Google's AI Turns News Reports into Flood Warnings for Vulnerable Regions

Google has developed an innovative flood prediction system by analyzing millions of news articles with its Gemini AI. The technology transforms qualitative reports into quantitative data, creating early warnings for areas lacking traditional weather monitoring. Already implemented in 150 countries, this approach marks a breakthrough in using language models for disaster prevention while addressing global inequality in weather forecasting capabilities.

March 13, 2026
AI innovationdisaster preventionclimate technology
News

NVIDIA's Nemotron 3 Super shakes up AI with open-source power rivaling top models

NVIDIA has unleashed Nemotron 3 Super, a groundbreaking open-source AI model that's turning heads with performance nearly matching premium closed-source alternatives like GPT-5.4. This 120-billion-parameter powerhouse combines innovative architecture with practical efficiency, delivering triple the reasoning speed while maintaining impressive accuracy. Already adopted by major tech players, it could democratize access to high-performance AI tools.

March 12, 2026
AI developmentOpen-source technologyNVIDIA