Add comprehensive summary of Gemini CLI capabilities - #1

This PR addresses the request to summarize what the Gemini CLI can do by creating a comprehensive capabilities document.

What's Added

Added GEMINI_CLI_CAPABILITIES.md - a detailed summary document that covers all aspects of the Gemini CLI's functionality, including:

Core Capabilities

• AI Model Access: Direct integration with Gemini 2.5 Pro/Flash models with 1M token context
• Multiple Authentication: OAuth, API key, and Vertex AI options
• Interactive & Non-Interactive Modes: Both terminal sessions and scripting support

Built-in Tools & Features

• File System Operations: Read/write files, directory listing, multi-file processing, smart editing
• Web Integration: Google Search grounding, URL fetching (up to 20 URLs), content processing
• Shell Integration: Command execution with safety confirmations, dedicated shell mode
• Memory & Context Management: Session persistence, hierarchical memory, custom context files

Advanced Features

• MCP (Model Context Protocol) Support: Extensibility via external servers and custom tools
• Conversation Management: Checkpointing, state branching, session compression
• Multimodal Capabilities: Image/PDF processing, code generation from sketches
• Custom Command System: TOML-based reusable commands with dynamic content injection

Developer Experience

• Rich Command System: Slash commands (/), content injection (@), shell passthrough (!)
• GitHub Integration: Automated PR reviews, issue triage, workflow automation
• Security Features: Sandboxing, confirmation prompts, workspace boundaries
• Terminal-First Design: Vim mode, themes, keyboard shortcuts

Purpose

This document serves as a comprehensive reference for understanding the full scope of Gemini CLI's capabilities, from basic AI interactions to advanced automation workflows. It's structured to help both new users understand what's possible and existing users discover features they might not be aware of.

The summary demonstrates that Gemini CLI is not just a simple AI chat interface, but a sophisticated development environment that bridges AI capabilities with practical development workflows.

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.