refactor(ui-design): shift from automatic capture to user-driven input model

Major Changes:
- Remove MCP-based automatic screenshot capture dependency
- Remove automatic code extraction mechanisms
- Shift to user-provided images and prompts as primary input

animation-extract.md:
- Remove MCP chrome-devtools integration
- Change from URL-based to image-based inference
- Use AI analysis of images instead of CSS extraction
- Update parameters: --urls → --images (glob pattern)

explore-auto.md:
- Fix critical parameter passing bug in animation-extract phase
- Add --images and --prompt parameter passing to animation-extract
- Ensure consistent context propagation across all extraction phases

imitate-auto.md:
- Add --refine parameter to all extraction commands (style, animation, layout)
- Add --images parameter passing to animation-extract phase
- Add --prompt parameter passing to animation-extract phase
- Align with refinement-focused design intent

Parameter Transmission Matrix (Fixed):
- style-extract: --images ✓ --prompt ✓ --refine ✓ (imitate only)
- animation-extract: --images ✓ --prompt ✓ --refine ✓ (imitate only)
- layout-extract: --images ✓ --prompt ✓ --refine ✓ (imitate only)

Deprecation Notice:
- capture.md and explore-layers.md will be removed in future versions
- Workflows now rely on user-provided inputs instead of automated capture

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

.claude/commands/workflow/ui-design/imitate-auto.md

@@ -298,7 +298,7 @@ ELSE:
             extraction_prompt = f"{--prompt} (supplement code-imported tokens)"
         command_parts.append(f"--prompt \"{extraction_prompt}\"")
 
-    command_parts.extend(["--variants 1", "--interactive"])
+    command_parts.extend(["--variants 1", "--refine", "--interactive"])
 
     extract_command = " ".join(command_parts)
     SlashCommand(extract_command)
@@ -315,7 +315,19 @@ IF skip_animation:
     REPORT: "✅ Phase 2.3: Animation (Using Code Import)"
 ELSE:
     REPORT: "🚀 Phase 2.3: Animation Extraction"
-    animation_extract_command = f"/workflow:ui-design:animation-extract --design-id \"{design_id}\" --interactive"
+
+    # Build command with available inputs
+    command_parts = [f"/workflow:ui-design:animation-extract --design-id \"{design_id}\""]
+
+    IF --images:
+        command_parts.append(f"--images \"{--images}\"")
+
+    IF --prompt:
+        command_parts.append(f"--prompt \"{--prompt}\"")
+
+    command_parts.extend(["--refine", "--interactive"])
+
+    animation_extract_command = " ".join(command_parts)
     SlashCommand(animation_extract_command)
 
 TodoWrite(mark_completed: "Extract animation", mark_in_progress: "Extract layout")
@@ -342,7 +354,7 @@ ELSE:
 
     # Default target if not specified
     command_parts.append("--targets \"home\"")
-    command_parts.extend(["--variants 1", "--interactive"])
+    command_parts.extend(["--variants 1", "--refine", "--interactive"])
 
     layout_extract_command = " ".join(command_parts)
     SlashCommand(layout_extract_command)