- Add VCR patch for binary request bodies (base64 encoding fallback)
- Add generate_filename() utility for UUID-based filenames with extension
- Add OpenAIResponsesFormatter for Responses API (input_image, input_file)
- Fix OpenAI uploader to use 'vision' purpose for images
- Fix Anthropic uploader to use tuple format (filename, content, content_type)
- Add TextConstraints and text support for Gemini
- Add file_id upload integration tests for Anthropic and OpenAI Responses API
- Add format_text_content override for Bedrock's {"text": ...} format
- Handle pre-formatted list content in _format_messages_for_converse
- Update Bedrock tests to use Claude 3 Haiku for on-demand availability
- Add VCR cassettes for Bedrock multimodal tests
Record VCR cassettes for OpenAI, Anthropic, Azure, and LiteLLM
multimodal tests. Gemini and Bedrock tests remain but cassettes
will be generated when credentials are available.
- Make crewai_files an optional dependency with graceful fallbacks
- Move file formatting from executor to LLM layer (_process_message_files)
- Add files field to LLMMessage type for cleaner message passing
- Add cache_control to Anthropic content blocks for prompt caching
- Clean up formatters: static methods for OpenAI/Gemini, proper error handling
- Remove unused ContentFormatter protocol
- Move test fixtures to lib/crewai-files/tests/fixtures
- Add Azure and Bedrock multimodal integration tests
- Fix mypy errors in crew_agent_executor.py
- Add format_multimodal_content() to all LLM providers
- Support inline base64 and file reference formats
- Add FileResolver integration for upload caching
- Add module exports for files package
