Memory actions work

src/models/system3.prompt

@@ -34,7 +34,7 @@ If you're over this budget, you can no longer create new tools. In case this hap
 </Info>
 
 <Info>
-If user shares something important or sensitive, you should usee the UpdateMemory tool to store this information. Include details such as the user's preferences, how you solved errors, and any other relevant context that could help in future interactions. This will help you provide better assistance in the future.
+If user shares something important or sensitive, you should usee the MemoryManager tool to store this information. Include details such as the user's preferences, how you solved errors, and any other relevant context that could help in future interactions. This will help you provide better assistance in the future.
 </Info>
 
 Here's a set of rules you must follow:
@@ -123,25 +123,42 @@ When you go over the resource budget, you must carefully evaluate which agent is
 </Rule>
 
 <Rule>
-Whenever you encounter an error — in logic, execution, tool invocation, or agent behavior — and resolve it, you must invoke the UpdateMemory tool immediately. The stored memory must include: 
-(1) the nature of the error, 
-(2) how it was identified, 
-(3) the correction applied, and 
-(4) the conversational or technical context in which it occurred.
+When you encounter and successfully fix an error — including logic flaws, failed tool invocations, or agent misbehavior — you must invoke the MemoryManager tool with:
+- action: "add_memory"
+- memory: A detailed string describing:
+  • The original error,
+  • How it was detected,
+  • The applied fix,
+  • And the context of the issue.
+
+This ensures persistent tracking of learning and system evolution.
+</Rule>
+
+<Rule>
+Whenever the user provides feedback, corrections, or clarification — whether explicitly or implicitly — you must invoke the MemoryManager tool with:
+- action: "add_memory"
+- memory: A string summarizing:
+  • The user’s input verbatim,
+  • The assistant’s original behavior that was corrected,
+  • The updated behavior or lesson learned,
+  • And any inferred user preferences or tone adjustments.
+
+This enables personalized, adaptive responses over time.
 </Rule>
 
 <Rule>
-Whenever the user provides corrections, preferences, or feedback — whether direct or implied — you must invoke the UpdateMemory tool to store this information. The stored memory must include:
-(1) the user’s exact correction or feedback,
-(2) the assistant's original output that was corrected or improved,
-(3) any inferred preferences, communication style, tone, or priorities the user reveals.
+If the user shares enduring personal context — such as biographical facts, technical expertise, long-term goals, emotional tendencies, or preferred interaction style — you must invoke the MemoryManager tool with:
+- action: "add_memory"
+- memory: A string encoding this information for future reference.
+
+This ensures continuity and personalization across sessions.
 </Rule>
 
 <Rule>
-If the user shares any information that is relevant to their identity, values, goals, recurring interests, working style, emotional state, or long-term projects, you must store it using the UpdateMemory tool. Ensure this includes:
-- Important biographical or academic details
-- Technical preferences or frequently used frameworks/tools
-- Emotional patterns or motivational triggers
-- Career objectives, research themes, or ongoing concerns
-- Personal principles, values, or recurring philosophical stances
+When the user explicitly requests deletion of a memory, or when a previously stored memory becomes obsolete, invalid, or counterproductive, you must invoke the MemoryManager tool with:
+- action: "delete_memory"
+- index: The index of the memory to remove (as returned from a prior "get_memory" action).
+
+Always confirm that the index is valid before deletion. If uncertain, retrieve the memory list first.
 </Rule>
+

src/tools/default_tools/memory_manager.py

@@ -0,0 +1,96 @@
+
+__all__ = ['MemoryManager']
+
+import json
+
+
+class MemoryManager():
+    dependencies = []
+
+    inputSchema = {
+        "name": "MemoryManager",
+        "description": "Updates, retrieves, or deletes the memory of an AI agent. The memory is stored in a JSON file.",
+        "parameters": {
+            "type": "object",
+            "properties":{
+                "action": {
+                    "type": "string",
+                    "enum": ["add_memory", "get_memory", "delete_memory"],
+                    "description": "The action to perform: add_memory, get_memory, or delete_memory.",
+                },
+                "memory": {
+                    "type": "string",
+                    "description": "The memory to add. Required for 'add_memory' action.",
+                },
+                "index": {
+                    "type": "integer",
+                    "description": "The index of the memory to delete. Required for 'delete_memory' action.",
+                },
+            },
+            "required": ["action"],
+        },
+    }
+    
+    def get_memories(self):
+        # load the memory from src/data/memory.json
+        try:
+            with open("src/data/memory.json", "r") as f:
+                memory_list = json.load(f)
+        except FileNotFoundError:
+            memory_list = []
+        except json.JSONDecodeError:
+            memory_list = []
+        return memory_list
+
+    def update_memories(self, memories):
+        # save the memory to src/data/memory.json
+        with open("src/data/memory.json", "w") as f:
+            json.dump(memories, f)
+
+
+    def run(self, **kwargs):
+        # save it to src/data/memory.json
+        action = kwargs.get("action")
+        memory = kwargs.get("memory")
+        index = kwargs.get("index")
+        memories = self.get_memories()
+        if action == "get_memory":
+            return {
+                "status": "success",
+                "message": "Memory retrieved successfully",
+                "output": memories
+            }
+        if action == "add_memory":
+            if memory is None:
+                return {
+                    "status": "error",
+                    "message": "Memory is required for add_memory action",
+                    "output": None
+                }
+            memories.append(memory)
+            self.update_memories(memories)
+            return {
+                "status": "success",
+                "message": "Memory created successfully",
+                "output": None
+            }
+        if action == "delete_memory":
+            if index is None:
+                return {
+                    "status": "error",
+                    "message": "Index is required for delete_memory action",
+                    "output": None
+                }
+            if index < 0 or index >= len(memories):
+                return {
+                    "status": "error",
+                    "message": "Index out of range",
+                    "output": None
+                }
+            memories.pop(index)
+            self.update_memories(memories)
+            return {
+                "status": "success",
+                "message": "Memory deleted successfully",
+                "output": None
+            }

src/tools/default_tools/update_memory.py

@@ -1,55 +0,0 @@
-
-__all__ = ['UpdateMemory']
-
-import json
-
-
-class UpdateMemory():
-    dependencies = []
-
-    inputSchema = {
-        "name": "UpdateMemory",
-        "description": "Updates the memory of the AI agent. This tool is used to update the memory of the AI agent with new information.",
-        "parameters": {
-            "type": "object",
-            "properties":{
-                "memory": {
-                    "type": "string",
-                    "description": "The new memory to be added to the AI agent's memory.",
-                }
-            },
-            "required": ["memory"],
-        },
-    }
-
-
-    def run(self, **kwargs):
-        # save it to src/data/memory.json
-        memory = kwargs.get("memory")
-        if not memory:
-            return {
-                "status": "error",
-                "message": "Memory is required",
-                "output": None
-            }
-        # add the memory to the memory.json file which is list of strings
-        # create the file if it does not exist
-        try:
-            with open("src/data/memory.json", "r") as f:
-                memory_list = json.load(f)
-        except FileNotFoundError:
-            memory_list = []
-        except json.JSONDecodeError:
-            memory_list = []
-        
-        # add the new memory to the list
-        memory_list.append(memory)
-        # save the list to the file
-        with open("src/data/memory.json", "w") as f:
-            json.dump(memory_list, f)
-        # return the list of memories
-        return {
-            "status": "success",
-            "message": "Memory updated successfully",
-            "output": memory_list
-        }