Adding a sequence diagram

Author: jabrena

README.md

@@ -11,6 +11,8 @@ Modern Java IDEs, such as **Cursor AI**, provide ways to customize how the `Agen
 
 A system prompt is a set of instructions given to an AI model that defines how it should behave, what role it should take on, and what guidelines it should follow when responding to users. Think of it as the "operating manual" that shapes the AI's personality, capabilities, and boundaries.
 
+[Further information](./docs/cursor-interaction-sequence.png)
+
 ## How to use the Cursor rules?
 
 Using the Cursor rules is straightforward: simply `drag and drop` the cursor rule that you need into the chat textbox where you are typing your `User prompt`.

docs/cursor-interaction-sequence.png

@@ -0,0 +1,128 @@
+@startuml
+!define RECTANGLE class
+
+skinparam backgroundColor white
+skinparam handwritten false
+skinparam monochrome false
+skinparam shadowing false
+
+title Cursor AI Assistant Interaction Flow (Enhanced)
+
+actor "Software Engineer" as SE
+participant "Cursor IDE" as Cursor
+participant "System Prompt\nProcessor" as SystemPrompt
+participant "Context Manager" as Context
+participant "Memory System" as Memory
+participant "AI Model\n(Claude/GPT-4/etc.)" as AI
+participant "Tool Executor" as Tools
+participant "Web Search" as Web
+participant "Git Integration" as Git
+participant "Linter/Error\nChecker" as Linter
+
+== Prompt Initiation ==
+SE -> Cursor: Enter prompt/query in chat
+activate Cursor
+
+== Context & Memory Gathering ==
+Cursor -> Context: Gather current context
+activate Context
+Context -> Context: Collect open files\ncursor position\nproject structure
+Context -> Context: Analyze recent edits\nand file history
+Context --> Cursor: Return context data
+deactivate Context
+
+Cursor -> Memory: Retrieve relevant memories
+activate Memory
+Memory -> Memory: Search for related\nconversation history
+Memory --> Cursor: Return memory context
+deactivate Memory
+
+== System Prompt Processing ==
+Cursor -> SystemPrompt: Process user query with context
+activate SystemPrompt
+SystemPrompt -> SystemPrompt: Apply workspace rules\nand coding guidelines
+SystemPrompt -> SystemPrompt: Format system instructions\n(tool usage, constraints)
+SystemPrompt -> SystemPrompt: Combine user query +\nsystem prompt + context + memories
+SystemPrompt --> Cursor: Return formatted prompt
+deactivate SystemPrompt
+
+== AI Model Selection & Interaction ==
+Cursor -> AI: Send complete prompt with:\n- System instructions\n- User query\n- Context data\n- Available tools\n- Memory context
+activate AI
+
+AI -> AI: Select appropriate model\n(Claude Sonnet, GPT-4, etc.)
+AI -> AI: Process prompt and determine\nrequired actions
+
+== Parallel Tool Execution ==
+alt Multiple Tools Required
+    AI -> Tools: Execute file operations\n(read_file, edit_file, etc.)
+    activate Tools
+    Tools --> AI: Return file results
+    deactivate Tools
+    
+    AI -> Web: Search for real-time info
+    activate Web
+    Web --> AI: Return search results
+    deactivate Web
+    
+    AI -> Git: Fetch PR/commit info
+    activate Git
+    Git --> AI: Return Git data
+    deactivate Git
+    
+    AI -> AI: Process all tool results\nand synthesize response
+end
+
+== Error Detection & Correction ==
+alt Code Changes Made
+    AI -> Linter: Check for linting errors
+    activate Linter
+    Linter --> AI: Return error status
+    deactivate Linter
+    
+    alt Errors Found
+        AI -> AI: Plan error corrections
+        AI -> Tools: Apply fixes (up to 3 iterations)
+        activate Tools
+        Tools --> AI: Confirm fixes applied
+        deactivate Tools
+    end
+end
+
+== Memory Management ==
+AI -> Memory: Update/create memories\nif new insights gained
+activate Memory
+Memory -> Memory: Store relevant information\nfor future conversations
+deactivate Memory
+
+AI --> Cursor: Return AI response with:\n- Code suggestions\n- Explanations\n- Next steps\n- Memory updates
+deactivate AI
+
+== Response Delivery ==
+Cursor -> Cursor: Format response for display
+Cursor -> SE: Display AI response in chat
+deactivate Cursor
+
+== Code Application & Iteration ==
+alt Code Changes Suggested
+    SE -> Cursor: Accept/Apply code changes
+    activate Cursor
+    Cursor -> Cursor: Apply edits to files
+    Cursor -> SE: Confirm changes applied
+    deactivate Cursor
+    
+    alt Further Clarification Needed
+        SE -> Cursor: Follow-up questions/refinements
+        note right: Multi-turn conversation continues\nwith maintained context and memory
+        activate Cursor
+        Cursor -> Context: Update context with new changes
+        Cursor -> AI: Process follow-up with updated context
+        note over AI: Iterative refinement cycle\ncan repeat multiple times
+        deactivate Cursor
+    end
+end
+
+== Background Processes ==
+note over Tools: Some operations can run\nin background (long-running tasks)
+
+@enduml