Updated model defaults and the agent runtime, and published the component as an npm package:

• Updated the Model default on Create Chat Completion and Create Response to gpt-6-luna, replacing gpt-3.5-turbo, which OpenAI has shut down, and the retiring gpt-4.1
• Updated the Model default on Agent: Create Image Generation Tool to gpt-image-2.5-flare, replacing gpt-image-1, which OpenAI has shut down
• Updated the Model dropdown to offer gpt-6-astra, gpt-6.1-sol, gpt-6.1-terra and gpt-6-luna in place of the retiring GPT-5 and GPT-4.1 snapshots
• Updated the OpenAI Agents SDK behind the agent actions; a run that Agent: Run paused for approval before this version can no longer be continued with Agent: Resume Run and has to be started again, and model calls made after a resume now count toward Max Turns
• Published the component as an npm package, so code-native integrations can install it and call its actions, triggers and data sources from code with TypeScript types
• Updated the component description to better summarize what the component does

Updated spectral version

Added global debug support across all actions for improved troubleshooting

Various modernizations and documentation updates

Added inline data source for files to enable dynamic dropdown selection

Added file operations and enhanced agent capabilities:

• New file management actions: Upload File, Download File, and Delete File for use with OpenAI
• Tool Flow Trigger to enable flows as callable AI agent tools
• File attachment support in agent runs for processing PDFs, images, and documents
• Enhanced agent branching with file IDs and MCP server integration
• Improved agent configuration with custom tools and MCP servers
• Comprehensive test coverage for new file operations

Added AI Agents functionality with support for building agents, custom and OpenAI hosted tools, and MCP server connections

Modernized OpenAI component with updated libraries and upstream API changes