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

uixt/ai/parser_default.go

@@ -53,7 +53,7 @@ func (p *JSONContentParser) Parse(content string, size types.Size) (*PlanningRes
 	// Define a temporary struct to parse the expected JSON format
 	var jsonResponse struct {
 		Actions []Action `json:"actions"`
-		Summary string   `json:"summary"`
+		Thought string   `json:"thought"`
 		Error   string   `json:"error"`
 	}
 
@@ -95,9 +95,8 @@ func (p *JSONContentParser) Parse(content string, size types.Size) (*PlanningRes
 	toolCalls := convertActionsToToolCalls(normalizedActions, p.actionMapping)
 
 	return &PlanningResult{
-		ToolCalls:     toolCalls,
-		ActionSummary: jsonResponse.Summary,
-		Thought:       jsonResponse.Summary,
-		Content:       content,
+		ToolCalls: toolCalls,
+		Thought:   jsonResponse.Thought,
+		Content:   content,
 	}, nil
 }

uixt/ai/parser_ui_tars.go

@@ -52,10 +52,9 @@ func (p *UITARSContentParser) Parse(content string, size types.Size) (*PlanningR
 	toolCalls := convertActionsToToolCalls(actions, p.actionMapping)
 
 	return &PlanningResult{
-		ToolCalls:     toolCalls,
-		ActionSummary: thought,
-		Thought:       thought,
-		Content:       content,
+		ToolCalls: toolCalls,
+		Thought:   thought,
+		Content:   content,
 	}, nil
 }
 

uixt/ai/planner.go

@@ -27,11 +27,10 @@ type PlanningOptions struct {
 
 // PlanningResult represents the result of planning
 type PlanningResult struct {
-	ToolCalls     []schema.ToolCall `json:"tool_calls"`
-	ActionSummary string            `json:"summary"`
-	Thought       string            `json:"thought"`
-	Content       string            `json:"content"` // original content from model
-	Error         string            `json:"error,omitempty"`
+	ToolCalls []schema.ToolCall `json:"tool_calls"`
+	Thought   string            `json:"thought"`
+	Content   string            `json:"content"` // original content from model
+	Error     string            `json:"error,omitempty"`
 }
 
 func NewPlanner(ctx context.Context, modelConfig *ModelConfig) (*Planner, error) {
@@ -125,8 +124,8 @@ func (p *Planner) Call(ctx context.Context, opts *PlanningOptions) (*PlanningRes
 		})
 		// history will be appended with tool calls execution result
 		result := &PlanningResult{
-			ToolCalls:     message.ToolCalls,
-			ActionSummary: message.Content,
+			ToolCalls: message.ToolCalls,
+			Thought:   message.Content,
 		}
 		return result, nil
 	}
@@ -135,8 +134,8 @@ func (p *Planner) Call(ctx context.Context, opts *PlanningOptions) (*PlanningRes
 	result, err := p.parser.Parse(message.Content, opts.Size)
 	if err != nil {
 		result = &PlanningResult{
-			ActionSummary: message.Content,
-			Error:         err.Error(),
+			Thought: message.Content,
+			Error:   err.Error(),
 		}
 		log.Debug().Str("reason", err.Error()).Msg("parse content to actions failed")
 	}
@@ -147,7 +146,7 @@ func (p *Planner) Call(ctx context.Context, opts *PlanningOptions) (*PlanningRes
 	})
 
 	log.Info().
-		Interface("summary", result.ActionSummary).
+		Interface("thought", result.Thought).
 		Interface("tool_calls", result.ToolCalls).
 		Msg("get VLM planning result")
 	return result, nil

uixt/ai/planner_prompts.go

@@ -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
 }
 

uixt/android_test.go

@@ -25,7 +25,7 @@ func setupADBDriverExt(t *testing.T) *XTDriver {
 	require.Nil(t, err)
 	driverExt, err := NewXTDriver(driver,
 		option.WithCVService(option.CVServiceTypeVEDEM),
-		option.WithLLMService(option.LLMServiceTypeUITARS),
+		option.WithLLMService(option.LLMServiceTypeDoubaoVL),
 	)
 	require.Nil(t, err)
 	return driverExt

uixt/mcp_tools_utility.go

@@ -34,7 +34,8 @@ func (t *ToolSleep) Implement() server.ToolHandlerFunc {
 	return func(ctx context.Context, request mcp.CallToolRequest) (*mcp.CallToolResult, error) {
 		seconds, ok := request.Params.Arguments["seconds"]
 		if !ok {
-			return nil, fmt.Errorf("seconds parameter is required")
+			log.Warn().Msg("seconds parameter is required, using default value 5.0 seconds")
+			seconds = 5.0
 		}
 
 		// Sleep action logic