image-mcp-server

TermiVis 🚀

Visual intelligence for your terminal - One-click setup for image analysis in Claude Code

🎯 One-Click Installation

Get started with TermiVis in under 30 seconds:

Method 1: pipx (Recommended)

# Install TermiVis
pipx install termivls

# Setup with your API key
termivls setup --api-key YOUR_INTERNVL_API_KEY

Method 2: One-Line Installation

curl -sSL https://get.termivls.com | bash -s -- --api-key YOUR_INTERNVL_API_KEY

Method 3: Docker

docker run -e INTERNVL_API_KEY=YOUR_KEY -v ~/.config/claude-code:/app/.claude-config termivls/termivls

That's it! 🎉 TermiVis is now integrated with Claude Code.

✨ What You Get

• 🧠 Smart Image Analysis - AI-powered understanding of screenshots, diagrams, code, and designs
• 🔄 Natural Integration - Works seamlessly within Claude Code - no separate apps needed
• ⚡ Zero Configuration - One command setup, automatic Claude Code integration
• 🛠️ Developer-Focused - Optimized for debugging, UI review, code analysis, and design workflows

🚀 Instant Usage

After installation, just use Claude Code naturally:

# In Claude Code, simply ask:
"What's wrong with this error screenshot?" [attach image]
"How can I improve this UI design?" [attach design]
"What does this code do?" [attach code screenshot]
"Compare these two versions" [attach images]

No need to remember tool names or parameters - TermiVis automatically understands your intent!

📋 Available Commands

Command	Description
termivls setup	One-time setup with API key
termivls status	Check service health and configuration
termivls run	Run server manually (debugging)
termivls uninstall	Remove from Claude Code
termivls --help	Show all commands

🔧 Advanced Installation

From Source

git clone https://github.com/your-username/termivls
cd termivls
pip install -e .
termivls setup --api-key YOUR_KEY

Development Setup

git clone https://github.com/your-username/termivls
cd termivls
uv sync
source .venv/bin/activate
termivls setup --api-key YOUR_KEY

📊 System Requirements

• Python: 3.10 or higher
• OS: macOS, Linux, Windows (WSL)
• Claude Code: Latest version
• API Key: InternVL API access

🛠️ How It Works

1. Image Processing: Handles PNG, JPG, WEBP, GIF, BMP from files, URLs, or clipboard
2. Smart Analysis: Uses InternVL AI to understand images in development context
3. MCP Integration: Registers as a standard MCP server in Claude Code
4. Natural Interface: No technical commands - just describe what you need

🔍 Troubleshooting

Quick Health Check

termivls status

Common Issues

"Command not found"

# Restart terminal or:
source ~/.bashrc  # Linux/macOS
# Or add to PATH: ~/.local/bin

"API key invalid"

# Re-setup with correct key:
termivls setup --api-key YOUR_CORRECT_KEY --force

"Server not responding"

# Check logs and restart:
termivls run

Getting Help

• Status Check: termivls status
• Logs: termivls run (shows real-time logs)
• GitHub Issues: Report bugs

🎨 Use Cases

🐛 Debug Errors

"This Python error is confusing, what's the actual problem?"
[attach error screenshot]

🎨 Review UI/Design

"What's wrong with this interface from a UX perspective?"
[attach UI mockup]

💻 Analyze Code

"Can you explain this algorithm and suggest improvements?"
[attach code screenshot]

🔄 Compare Versions

"What changed between these two versions?"
[attach before/after images]

🚢 Deployment Options

Personal Use

pipx install termivls
termivls setup --api-key YOUR_KEY

Team/Organization

# Docker Compose
curl -O https://raw.githubusercontent.com/your-username/termivls/main/docker-compose.yml
INTERNVL_API_KEY=your_key docker-compose up -d

CI/CD Integration

- name: Install TermiVis
  run: pipx install termivls
- name: Setup TermiVis
  run: termivls setup --api-key ${{ secrets.INTERNVL_API_KEY }}

🤝 Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

# Setup development environment
git clone https://github.com/your-username/termivls
cd termivls
uv sync
source .venv/bin/activate

# Run tests
pytest tests/
python examples/test_basic_functionality.py

# Submit PR

📄 License

MIT License - see LICENSE for details.

🙏 Acknowledgments

• InternVL Team - For the powerful vision-language model
• Anthropic - For Claude Code and MCP protocol
• Python Community - For the amazing ecosystem

---

Made with ❤️ for developers who love efficient workflows

🌟 Star us on GitHub • 📚 Documentation • 🐛 Report Issues