- Implemented detection and handling for TodoWriteTool to route updates as plan entries instead of tool call events.
- Added sub-agent tool tracking for TaskTool, allowing for event emission and cleanup.
- Updated event listeners to manage sub-agent tool calls and approval requests effectively.
* fix: make token limits regex normalize e.g. `some-model-1.1` -> `some-model` while preserve e.g. `gpt-4.1` as-is.
* feat: update token limits regex for latest models `GLM-4.6`, `deepseek-v3.2-exp`.
* feat: add exact token limit number 202752 per the model config file for `GLM-4.6`.
* feat: Add Qwen3-VL-Plus token limits (256K input, 32K output)
- Added 256K input context window limit for Qwen3-VL-Plus model
- Updated output token limit from 8K to 32K for Qwen3-VL-Plus
- Added comprehensive tests for both input and output limits
As requested by Qwen maintainers for proper model support.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: enable high-res flag for qwen VL models
---------
Co-authored-by: Claude <noreply@anthropic.com>
- Added 256K input context window limit for Qwen3-VL-Plus model
- Updated output token limit from 8K to 32K for Qwen3-VL-Plus
- Added comprehensive tests for both input and output limits
As requested by Qwen maintainers for proper model support.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
* feat: add yolo mode support to auto vision model switch
* feat: add cli args & env variables for switch behavoir
* fix: use dedicated model names and settings
* docs: add vision model instructions
* fix: failed test case
* fix: setModel failure
