Penguin Code

    ____                        _          ______          __
   / __ \___  ____  ____ ___  _(_)___     / ____/___  ____/ /__
  / /_/ / _ \/ __ \/ __ `/ / / / / __ \   / /   / __ \/ __  / _ \
 / ____/  __/ / / / /_/ / /_/ / / / / /  / /___/ /_/ / /_/ /  __/
/_/    \___/_/ /_/\__, /\__,_/_/_/ /_/   \____/\____/\__,_/\___/
                 /____/

AI-powered coding assistant CLI and VS Code extension using Ollama

Penguin Tech Inc © 2025

Features

• 🤖 Multi-Agent System - ChatAgent orchestrates specialized Explorer/Executor agents
• 🔍 Multi-Engine Research - 5 search engines with MCP protocol support
• 🧠 Persistent Memory - mem0 integration for context across sessions
• 📚 Documentation RAG - Auto-detects your project's languages and libraries, fetches official documentation, and uses it for accurate, syntax-correct answers
• 🔌 MCP Integration - Extend with N8N, Flowise, and custom MCP servers
• 🌐 Client-Server Mode - gRPC server for remote Ollama and team deployments
• ⚡ GPU Optimized - Smart model switching for RTX 4060 Ti (8GB VRAM) or higher
• 🐧 Cross-Platform - Works on Linux, macOS, and Windows

Supported Languages

Language	Detection	Doc Sources
Python	pyproject.toml, requirements.txt, *.py	Official docs + PyPI libraries
JavaScript/TypeScript	package.json, tsconfig.json	MDN, npm packages
Go	go.mod, *.go	go.dev, pkg.go.dev
Rust	Cargo.toml, *.rs	docs.rs, crates.io
OpenTofu/Terraform	*.tf, *.tofu, .terraform.lock.hcl	OpenTofu docs, provider registries
Ansible	ansible.cfg, playbook.yml, requirements.yml	Ansible docs, Galaxy collections

Quick Start

# Install
pip install -e .
penguincode setup

# Pull required models
ollama pull llama3.2:3b qwen2.5-coder:7b nomic-embed-text

# Run
penguincode chat

VS Code Extension: Download VSIX from Releases

Server Mode (Team Deployment)

# Start gRPC server (connects to local Ollama)
python -m penguincode.server.main

# Or use Docker
docker compose up -d

# Connect from client
penguincode chat --server localhost:50051

See Architecture Documentation for remote deployment with TLS and authentication.

Documentation

• Usage Guide - Installation, configuration, and usage
• Configuration Reference - Complete config.yaml reference
• Architecture - Client-server architecture and deployment modes
• Agent Architecture - ChatAgent, Explorer, Executor, Planner
• Tool Support - Ollama models with native tool calling
• MCP Integration - Extend with N8N, Flowise, and custom servers
• Memory - Persistent memory with mem0 integration
• Documentation RAG - Project-aware documentation indexing
• Security - Authentication, TLS, and secure code generation
• Contributing - How to contribute

License

AGPL-3.0 - See LICENSE for details

Support: support.penguintech.io | Homepage: www.penguintech.io