Coc-llamautoma NPM

coc-llamautoma

AI-powered code generation and editing for coc.nvim, powered by Llamautoma. This extension provides seamless integration between your Vim/Neovim editor and the Llamautoma AI server, enabling advanced code generation, editing, and assistance features.

Features

AI Chat: Interactive chat with context-aware AI about your code
Smart Editing: AI-assisted code modifications across multiple files
Code Generation: Generate new files and code based on natural language descriptions
Workspace Sync: Keep AI context up-to-date with your workspace changes
LSP Integration: Full Language Server Protocol support
Real-time Updates: Streaming responses for immediate feedback
Safety Checks: Built-in safety measures for code modifications
Git-aware: Respects .gitignore patterns during synchronization
Multi-file Support: Handle changes across multiple files seamlessly
Autocompletion: AI-powered code completion suggestions

Prerequisites

Neovim >= 0.10.0 or Vim >= 9.0
coc.nvim >= 0.0.80
Bun
Ollama with required models:
- Default: qwen2.5-coder:7b (production)
- Testing: qwen2.5-coder:1.5b (faster for tests)

Installation

Via coc.nvim

:CocInstall coc-llamautoma

Manual Installation

cd ~/.config/coc/extensions
git clone https://github.com/dgpt/coc-llamautoma
cd coc-llamautoma
bun install
bun run build

Configuration

Basic Settings

Add these configurations to your coc-settings.json:

{
  "llamautoma.enable": true,
  "llamautoma.serverUrl": "http://localhost:3000",
  "llamautoma.timeout": 30000,
  "llamautoma.model": "qwen2.5-coder:7b",
  "llamautoma.autoSync": true,
  "llamautoma.syncOnSave": true,
  "llamautoma.syncIgnorePatterns": [
    "node_modules",
    "dist",
    "build",
    ".git"
  ],
  "llamautoma.maxFileSize": 1000000,
  "llamautoma.logLevel": "info"
}

Key Mappings

Add these to your (Neo)vim configuration:

" Chat with AI about current file
nmap <silent> <Leader>lc :CocCommand llama.chat<CR>

" Edit current file with AI
nmap <silent> <Leader>le :CocCommand llama.edit<CR>

" Generate new file with AI
nmap <silent> <Leader>ln :CocCommand llama.compose<CR>

" Sync workspace with AI
nmap <silent> <Leader>ls :CocCommand llama.sync<CR>

Commands

`:CocCommand llamautoma.chat`

Opens an interactive chat window with the AI
Context-aware of your current file and workspace
Supports code snippets and multi-line input
Use q to close the chat window
History is preserved per session

`:CocCommand llamautoma.edit`

Edit the current file with AI assistance
Supports multi-file edits
Shows preview of changes before applying
Safety checks for dangerous modifications
Undo/redo support for AI changes

`:CocCommand llamautoma.compose`

Generate new files with AI
Supports multiple file generation
Context-aware of your project structure
Automatic import management
Template-based generation

`:CocCommand llamautoma.sync`

Synchronize workspace with AI
Respects .gitignore patterns
Progress indicator for large workspaces
Selective sync support
Automatic periodic sync (if enabled)

Development

# Clone the repository
git clone https://github.com/yourusername/coc-llamautoma
cd coc-llamautoma

# Install dependencies
bun install

# Build the extension
bun run build

# Watch for changes
bun run watch

# Run tests
bun test

# Run tests with coverage
bun test --coverage

Project Structure

coc-llamautoma/
├── src/
│   ├── client/        # Llamautoma client implementation
│   ├── commands/      # Command implementations
│   ├── config/        # Configuration management
│   ├── handlers/      # LSP event handlers
│   └── utils/         # Utility functions
├── tests/
│   ├── unit/         # Unit tests
│   └── integration/  # Integration tests
└── package.json

Troubleshooting

Common Issues

Server Connection Failed
- Check if Llamautoma server is running
- Verify server URL in configuration
- Check network connectivity
Command Not Found
- Ensure extension is properly installed
- Run :CocList extensions to verify
- Try reinstalling the extension
Sync Issues
- Check workspace permissions
- Verify .gitignore patterns
- Check file size limits
Performance Issues
- Reduce workspace size
- Adjust sync settings
- Update to latest version