Computer Vision MCP Server

# Changelog All notable changes to this project are documented here. The format is based on Keep a Changelog, and this project adheres to Semantic Versioning where practical. ## [Unreleased] - Planned: additional CLI ergonomics and presets for local/remote testing. - Planned: Makefile targets for common workflows. ### Fixed - Local backend transformers compatibility: Replace non-existent `AutoModelForConditionalGeneration` with proper model classes (`Qwen2_5_VLForConditionalGeneration` for Qwen models) - JSON parsing for model outputs wrapped in markdown code blocks (```json...```) - Model loading fallback logic improved with better error handling ## [0.0.1] - 2025-09-14 ### Added - Metadata CLI backend overrides: `--ac-backend`, `--meta-vision-backend`, and `--local-model-id`. - Conditional OpenRouter key requirement in CLI (only when a selected step uses OpenRouter). - Documentation for testing local "double" with URL and fully local "triple" flows. ### Changed - `cli/image_metadata.py` help text clarified to cover both double and triple modes. ### Notes - Double mode still requires a text LLM via OpenRouter for metadata. Use triple mode with both backends set to `local` for fully local operation.