cognify-code 1.2.0

Your local AI-powered code assistant. Review, generate, search, and refactor code with an intelligent AI agent—all running locally with complete privacy.

🧠 Cognify AI

Code Cognition — Your AI-Powered Code Assistant

Features • VSCode Extension • Installation • Providers • Usage • Docs

---

A powerful CLI tool and VSCode extension that brings AI-powered code cognition to your development workflow. Review code, generate functions, search your codebase semantically, and refactor projects—with support for multiple LLM providers including local (Ollama) and cloud options with free tiers.

✨ Features

Feature	Description
🔍 Code Review	Analyze code for bugs, security issues, and style problems
⚡ Code Generation	Generate functions, classes, and tests from natural language
🔎 Semantic Search	Search your codebase using natural language queries
📝 AI File Editing	Edit files with natural language instructions
🔄 Multi-File Refactor	Refactor across multiple files at once
✏️ Symbol Renaming	Rename functions, classes, variables across your project
💬 Interactive Chat	Chat with AI about your code
📊 Codebase Indexing	Create searchable semantic index with RAG
🌐 Multi-Provider	Support for 6 LLM providers (local & cloud)
🖥️ VSCode Extension	Full IDE integration with sidebar chat
📁 File Operations	Create/modify/delete files directly from chat
🗂️ Multi-Session Chat	Multiple persistent chat sessions
🤖 Multi-Agent System	6 specialized AI agents with auto-selection
🧠 Project Intelligence	Auto-detect project type, frameworks, architecture
🐛 Debug Helper	Stack trace analysis and error debugging
🔍 Error Detection	Auto-detect errors in 7 languages
📱 Android/Termux	Mobile development support
⏹️ Stop Generation	Cancel AI responses mid-generation (NEW)
🧠 Smart Progress	Context-aware progress indicators (NEW)

🖥️ VSCode Extension

The Cognify AI VSCode extension brings all the power of Cognify directly into your IDE with a beautiful sidebar chat interface.

Extension Features (v1.2.0)

• ⚛️ Sidebar Chat Panel - Chat with AI directly in VSCode sidebar
• 📎 Add Context - Include code from your editor in conversations
• 🔍 Code Review - Review files with inline diagnostics
• ✨ Code Generation - Generate code from descriptions
• 💡 Code Explanation - Get explanations for selected code
• ✏️ AI Editing - Edit code with natural language
• 📁 File Operations - Create, modify, delete files from chat
• 🗂️ Multi-Session Chat - Multiple persistent chat sessions
• 🤖 Multi-Agent System - 6 specialized agents with auto-selection
• 🧠 Project Intelligence - Auto-detect project type & tech stack
• 🔄 Diff Engine - View and approve code changes before applying
• 🐛 Debug Helper - Stack trace analysis & error debugging
• 🔍 Error Detection - Auto-detect errors in 7 languages
• 📱 Android/Termux - Custom CLI path for mobile dev
• ⏹️ Stop Generation - Cancel AI responses mid-stream (NEW v1.2.0)
• 🧠 Smart Progress - Context-aware progress steps (NEW v1.2.0)

Quick Actions

• 📋 Review current file
• 💡 Explain selected code
• ✨ Suggest improvements
• 🧪 Generate tests

Install VSCode Extension

# Build from source
cd vscode-extension
npm install
npm run compile
npx vsce package --no-dependencies

# Install the extension
code --install-extension cognify-code-1.2.0.vsix

Or install via VSCode:

1. Open VSCode
2. Go to Extensions sidebar
3. Click "..." → "Install from VSIX..."
4. Select cognify-code-1.2.0.vsix

Using the Extension

1. Click the ⚛️ Cognify icon in the Activity Bar (left sidebar)
2. Use quick actions or type your question
3. Click "📎 Add Context" to include code from your editor
4. Press Enter or click Send

Keyboard Shortcuts:

Command	Mac	Windows/Linux
Review File	Cmd+Shift+R	Ctrl+Shift+R
Generate Code	Cmd+Shift+G	Ctrl+Shift+G
Explain Code	Cmd+Shift+E	Ctrl+Shift+E
Open Chat	Cmd+Shift+C	Ctrl+Shift+C

🤖 Supported Providers

Provider	Free Tier	API Key	Best For
Ollama	✅ Unlimited	❌ None	Privacy, offline use
Google AI	✅ Generous	✅ Required	1M+ token context
Groq	✅ 1000 req/day	✅ Required	Fastest inference
Cerebras	✅ Available	✅ Required	Fast inference
OpenRouter	✅ Free models	✅ Required	Model variety
OpenAI	❌ Paid only	✅ Required	GPT-4 quality

🚀 Quick Start

Option 1: Local with Ollama (No API Key)

# Install Ollama from https://ollama.ai
ollama pull deepseek-coder:6.7b
ollama serve

Option 2: Cloud with Free Tier

# Google AI Studio (1M token context, free)
export GOOGLE_API_KEY="your-key"  # Get from https://aistudio.google.com/apikey

# OR Groq (fastest inference, free tier)
export GROQ_API_KEY="your-key"    # Get from https://console.groq.com/keys

# OR OpenRouter (free models available)
export OPENROUTER_API_KEY="your-key"  # Get from https://openrouter.ai/keys

Installation

Quick Install

pip install cognify-code

Complete Installation Guide

For detailed setup including:

• Offline LLMs with Ollama (DeepSeek Coder, Qwen, CodeLlama)
• Cloud Providers (Groq, Google AI, Cerebras, OpenAI)
• VSCode Extension installation
• Android/Termux support
• Troubleshooting guide

👉 Complete Installation Guide

Install from Source

# Clone the repository
git clone https://github.com/akkssy/cognify-ai.git
cd cognify-ai

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install the package
pip install -e .

Verify Installation

# Check status and available providers
cognify status
cognify providers

📖 CLI Usage

Code Review

cognify review path/to/file.py
cognify review src/ --format json

Code Generation

cognify generate "binary search function" --language python
cognify generate "REST API client class" --mode class

Semantic Search

# First, index your codebase
cognify index .

# Then search
cognify search "error handling"

File Editing

cognify edit config.py "add logging to all functions" --preview

Interactive Chat

cognify chat

All Commands

cognify --help

⚙️ Configuration

Configuration is managed via config.yaml:

llm:
  provider: "ollama"
  model: "deepseek-coder:6.7b"
  base_url: "http://localhost:11434"
  temperature: 0.1
  max_tokens: 4096

Or use environment variables:

export AI_ASSISTANT_LLM_PROVIDER="groq"
export AI_ASSISTANT_LLM_MODEL="llama-3.3-70b-versatile"
export GROQ_API_KEY="your-key"

📁 Project Structure

cognify-ai/
├── src/ai_code_assistant/    # Core Python package
│   ├── cli.py                # Command-line interface
│   ├── providers/            # LLM providers
│   ├── reviewer/             # Code review
│   ├── generator/            # Code generation
│   ├── retrieval/            # Semantic search
│   ├── editor/               # AI file editing
│   └── chat/                 # Interactive chat
├── vscode-extension/         # VSCode Extension
│   ├── src/                  # Extension source
│   ├── images/               # Icons
│   └── package.json          # Extension manifest
├── tests/                    # Unit tests
├── docs/                     # Documentation
└── config.yaml               # Configuration

🧪 Testing

PYTHONPATH=src pytest tests/ -v

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

MIT License - see LICENSE for details.

---

Made with ❤️ for developers who want flexible AI-powered coding assistance