refactor: merge ActionSummary and Thought fields to eliminate duplication

- Remove redundant ActionSummary field from PlanningResult struct
- Update parsers to use unified Thought field instead of duplicate fields
- Modify chat interface to display Thought instead of ActionSummary
- Update planner logging to use thought instead of summary
- Adjust prompt templates to use thought field consistently
- Switch test LLM service from UI-TARS to DoubaoVL
- Add default parameter handling for sleep tool
This commit is contained in:
lilong.129
2025-06-05 14:19:09 +08:00
parent 0864f74021
commit 0add3231ff
8 changed files with 24 additions and 26 deletions
+3 -3
View File
@@ -98,7 +98,7 @@ Supporting actions:
Field description:
* The ` + "`start_box`" + ` and ` + "`end_box`" + ` fields represent the bounding box coordinates of the target element in 1000x1000 relative coordinate system.
* Use Chinese in log and summary fields.
* Use Chinese in log and thought fields.
Return in JSON format:
{
@@ -108,7 +108,7 @@ Return in JSON format:
"action_inputs": { ... }
}
],
"summary": "string", // Log what the next action you can do according to the screenshot and the instruction. Use Chinese.
"thought": "string", // Log what the next action you can do according to the screenshot and the instruction. Use Chinese.
"error": "string" | null, // Error messages about unexpected situations, if any. Use Chinese.
}
@@ -123,7 +123,7 @@ For example, when the instruction is "点击第二个帖子的作者头像", by
}
}
],
"summary": "点击第二个帖子的作者头像",
"thought": "点击第二个帖子的作者头像",
"error": null
}