Wingman AI is an open-source VSCode extension providing AI-powered coding assistance. It targets developers seeking integrated, agentic coding capabilities, offering features like code completion, refactoring, and documentation generation, with the benefit of local or cloud-based AI model integration.

How It Works

Wingman AI integrates with multiple large language model providers, including OpenAI, Anthropic, and the locally runnable Ollama. It supports the Model Context Protocol (MCP), enabling extensions with tools like Figma or browser interaction. This approach allows users to leverage powerful AI models directly within their VSCode environment, with the flexibility to choose between privacy-focused local models and cloud-based services, and to extend the assistant's functionality with various MCP-compatible tools.

Quick Start & Requirements

• Install via VSCode Extensions marketplace.
• Requires VSCode.
• Optional: Ollama, OpenAI API key, or Anthropic API key for AI model access.
• Documentation: GitHub

Highlighted Details

• Supports OpenAI, Anthropic, and local Ollama models.
• Integrates with Model Context Protocol (MCP) tools.
• Features agent-based chat, code completion, refactoring, and documentation generation.
• Allows highlighting text for focused AI assistance or providing requirements for autonomous feature implementation.

Maintenance & Community

• Active development on GitHub.
• Community engagement encouraged via GitHub issues.
• Roadmap and release notes available on GitHub.

Licensing & Compatibility

• License: MIT.
• Compatible with commercial and closed-source projects.

Limitations & Caveats

The extension is primarily focused on VSCode and its functionality is dependent on the chosen AI model provider's capabilities and availability. Advanced agentic features may require specific configurations or MCP tool integrations.