Release v1.3.1

- Improved osxcross detection in release build script
- MCP server now captures and includes stderr output in responses
- Fixed stdio handling for aid subprocess execution
- Progress bars and statistics are now visible in MCP client responses

chore: release AI Distiller v1.3.0 and migrate MCP to TypeScript SDK

Major changes:
- Bump version from 1.0.0 to 1.3.0 across all files
- Migrate MCP server from JavaScript to TypeScript SDK
  - Replace mcp-server.js with mcp-server-sdk.ts
  - Add TypeScript configuration and build process
  - Improve type safety and maintainability
- Enhance postinstall script with version checking
  - Now properly handles upgrades by comparing versions
  - Removes old binary if version mismatch detected
- Add publish.sh script for safer npm releases
  - Includes safety checks and confirmation prompts
  - Automates the entire build and publish process
- Dynamic version loading from package.json
  - No more hardcoded versions in multiple files
  - Single source of truth for versioning
- Add conditional debug logging (DEBUG=true environment variable)

Breaking changes:
- Node.js requirement increased from 14.0.0 to 18.0.0

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <[email protected]>

AI Distiller v1.0.0 - First Public Release

🎉 The first public release of AI Distiller is here\!

AI Distiller is a high-performance CLI tool that extracts essential code structure from large codebases, making them digestible for LLMs by removing unnecessary details while preserving semantic information.

- ⚡ **Blazingly fast**: Process 10MB codebase in <2 seconds
- 🌍 **12+ languages**: Python, TypeScript, JavaScript, Go, Java, C#, Rust, and more
- 🤖 **AI-optimized**: Multiple output formats designed for LLM consumption
- 🔧 **Flexible filtering**: Control exactly what to include/exclude
- 📦 **Zero dependencies**: All tree-sitter parsers bundled in single binary
- 🚀 **Parallel processing**: Near-linear speedup with CPU cores
- 🔌 **MCP Server**: Model Context Protocol support for AI agents

- Full language support via tree-sitter parsers
- Text, Markdown, JSON, JSONL, and XML output formats
- Granular visibility control (public/private/protected/internal)
- Smart project root detection with .aid/ centralization
- Git history analysis mode with AI prompts
- Comprehensive debugging system (3 levels)
- File pattern filtering with glob support
- stdin processing for pipeline integration

This release includes pre-built binaries for:
- Linux (AMD64, ARM64)
- macOS (Intel, Apple Silicon)
- Windows (AMD64)

All binaries include full language support with CGO-enabled tree-sitter parsers.

This is our first public release, and we're eager to hear your feedback\! Please report issues, suggest features, or share your use cases at:
https://github.com/janreges/ai-distiller/issues

Thank you to all early testers and contributors who helped shape this tool\!