memorisdk 2.1.1__tar.gz → 2.3.1__tar.gz

{memorisdk-2.1.1 → memorisdk-2.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: memorisdk
-Version: 2.1.1
+Version: 2.3.1
 Summary: The Open-Source Memory Layer for AI Agents & Multi-Agent Systems
 Author-email: GibsonAI Team <noc@gibsonai.com>
 License: Apache-2.0
@@ -95,7 +95,7 @@ Requires-Dist: pandas>=2.0.0; extra == "all"
 Requires-Dist: plotly>=5.17.0; extra == "all"
 Dynamic: license-file
 
-[![GibsonAI](https://github.com/user-attachments/assets/878e341b-5a93-4489-a398-abeca91b6b11)](https://gibsonai.com/)
+[![Memori Labs](https://s3.us-east-1.amazonaws.com/images.memorilabs.ai/banner.png)](https://memorilabs.ai/)
 
 # memori
 
@@ -110,7 +110,7 @@ Dynamic: license-file
 <p align="center">
   <a href="https://memori.gibsonai.com/docs">Learn more</a>
   ·
-  <a href="https://www.gibsonai.com/discord">Join Discord</a>
+  <a href="https://discord.gg/abD4eGym6v">Join Discord</a>
 </p>
 
 <p align="center">
@@ -120,14 +120,20 @@ Dynamic: license-file
   <a href="https://pepy.tech/projects/memorisdk">
     <img src="https://static.pepy.tech/badge/memorisdk" alt="Downloads">
   </a>
-  <a href="https://opensource.org/licenses/MIT">
-    <img src="https://img.shields.io/badge/License-MIT-yellow.svg" alt="License: MIT">
+  <a href="https://opensource.org/license/apache-2-0">
+    <img src="https://img.shields.io/badge/license-Apache%20License%202.0-blue" alt="License: Apache 2.0">
   </a>
   <a href="https://www.python.org/downloads/">
     <img src="https://img.shields.io/badge/python-3.8+-blue.svg" alt="Python 3.8+">
   </a>
 </p>
 
+<p align="center">
+  <a href="https://github.com/GibsonAI/memori/stargazers">
+    <img src="https://img.shields.io/badge/⭐%20Give%20a%20Star-Support%20the%20project-orange?style=for-the-badge" alt="Give a Star">
+  </a>
+</p>
+
 ---
 
 ## What is Memori
@@ -206,6 +212,8 @@ print("\n💡 Notice: Memori automatically knows about your FastAPI Python proje
 
 ---
 
+⭐️ **Enjoying Memori?** Give us a star to support open development
+
 > By default, Memori uses in-memory SQLite database. Get **FREE** serverless database instance in [GibsonAI](https://app.gibsonai.com/signup) platform.
 
 **🚀 Ready to explore more?**
@@ -326,6 +334,7 @@ memori = Memori(
     database_connect="sqlite:///my_memory.db",
     template="basic", 
     conscious_ingest=True,  # One-shot context injection
+    conscious_memory_limit=100,  # Must be an integer between 1 and 500
     openai_api_key="sk-..."
 )
 
@@ -482,6 +491,7 @@ Memori works seamlessly with popular AI frameworks:
 | [Agno](./examples/integrations/agno_example.py) | Memory-enhanced agent framework integration with persistent conversations | Simple chat agent with memory search |
 | [AWS Strands](./examples/integrations/aws_strands_example.py) | Professional development coach with Strands SDK and persistent memory | Career coaching agent with goal tracking |
 | [Azure AI Foundry](./examples/integrations/azure_ai_foundry_example.py) | Azure AI Foundry agents with persistent memory across conversations | Enterprise AI agents with Azure integration |
+| [AutoGen](./examples/integrations/autogen_example.py) |  Multi-agent group chat memory recording | Agent chats with memory integration |
 | [CamelAI](./examples/integrations/camelai_example.py) | Multi-agent communication framework with automatic memory recording and retrieval | Memory-enhanced chat agents with conversation continuity |
 | [CrewAI](./examples/integrations/crewai_example.py) | Multi-agent system with shared memory across agent interactions | Collaborative agents with memory |
 | [Digital Ocean AI](./examples/integrations/digital_ocean_example.py) | Memory-enhanced customer support using Digital Ocean's AI platform | Customer support assistant with conversation history |
@@ -502,11 +512,15 @@ Explore Memori's capabilities through these interactive demonstrations:
 ## 🤝 Contributing
 
 - See [CONTRIBUTING.md](./CONTRIBUTING.md) for development setup and guidelines.
-- Community: [Discord](https://www.gibsonai.com/discord)
+- Community: [Discord](https://discord.gg/abD4eGym6v)
+
+## ⭐️ Star us on GitHub to support the project 
+
+[![Star History Chart](https://api.star-history.com/svg?repos=GibsonAI/memori&type=date&legend=top-left)](https://www.star-history.com/#GibsonAI/memori&type=date&legend=top-left)
 
 ## 📄 License
 
-MIT License - see [LICENSE](./LICENSE) for details.
+Apache 2.0 License - see [LICENSE](./LICENSE) for details.
 
 ---
 

{memorisdk-2.1.1 → memorisdk-2.3.1}/README.md

@@ -1,4 +1,4 @@
-[![GibsonAI](https://github.com/user-attachments/assets/878e341b-5a93-4489-a398-abeca91b6b11)](https://gibsonai.com/)
+[![Memori Labs](https://s3.us-east-1.amazonaws.com/images.memorilabs.ai/banner.png)](https://memorilabs.ai/)
 
 # memori
 
@@ -13,7 +13,7 @@
 <p align="center">
   <a href="https://memori.gibsonai.com/docs">Learn more</a>
   ·
-  <a href="https://www.gibsonai.com/discord">Join Discord</a>
+  <a href="https://discord.gg/abD4eGym6v">Join Discord</a>
 </p>
 
 <p align="center">
@@ -23,14 +23,20 @@
   <a href="https://pepy.tech/projects/memorisdk">
     <img src="https://static.pepy.tech/badge/memorisdk" alt="Downloads">
   </a>
-  <a href="https://opensource.org/licenses/MIT">
-    <img src="https://img.shields.io/badge/License-MIT-yellow.svg" alt="License: MIT">
+  <a href="https://opensource.org/license/apache-2-0">
+    <img src="https://img.shields.io/badge/license-Apache%20License%202.0-blue" alt="License: Apache 2.0">
   </a>
   <a href="https://www.python.org/downloads/">
     <img src="https://img.shields.io/badge/python-3.8+-blue.svg" alt="Python 3.8+">
   </a>
 </p>
 
+<p align="center">
+  <a href="https://github.com/GibsonAI/memori/stargazers">
+    <img src="https://img.shields.io/badge/⭐%20Give%20a%20Star-Support%20the%20project-orange?style=for-the-badge" alt="Give a Star">
+  </a>
+</p>
+
 ---
 
 ## What is Memori
@@ -109,6 +115,8 @@ print("\n💡 Notice: Memori automatically knows about your FastAPI Python proje
 
 ---
 
+⭐️ **Enjoying Memori?** Give us a star to support open development
+
 > By default, Memori uses in-memory SQLite database. Get **FREE** serverless database instance in [GibsonAI](https://app.gibsonai.com/signup) platform.
 
 **🚀 Ready to explore more?**
@@ -229,6 +237,7 @@ memori = Memori(
     database_connect="sqlite:///my_memory.db",
     template="basic", 
     conscious_ingest=True,  # One-shot context injection
+    conscious_memory_limit=100,  # Must be an integer between 1 and 500
     openai_api_key="sk-..."
 )
 
@@ -385,6 +394,7 @@ Memori works seamlessly with popular AI frameworks:
 | [Agno](./examples/integrations/agno_example.py) | Memory-enhanced agent framework integration with persistent conversations | Simple chat agent with memory search |
 | [AWS Strands](./examples/integrations/aws_strands_example.py) | Professional development coach with Strands SDK and persistent memory | Career coaching agent with goal tracking |
 | [Azure AI Foundry](./examples/integrations/azure_ai_foundry_example.py) | Azure AI Foundry agents with persistent memory across conversations | Enterprise AI agents with Azure integration |
+| [AutoGen](./examples/integrations/autogen_example.py) |  Multi-agent group chat memory recording | Agent chats with memory integration |
 | [CamelAI](./examples/integrations/camelai_example.py) | Multi-agent communication framework with automatic memory recording and retrieval | Memory-enhanced chat agents with conversation continuity |
 | [CrewAI](./examples/integrations/crewai_example.py) | Multi-agent system with shared memory across agent interactions | Collaborative agents with memory |
 | [Digital Ocean AI](./examples/integrations/digital_ocean_example.py) | Memory-enhanced customer support using Digital Ocean's AI platform | Customer support assistant with conversation history |
@@ -405,11 +415,15 @@ Explore Memori's capabilities through these interactive demonstrations:
 ## 🤝 Contributing
 
 - See [CONTRIBUTING.md](./CONTRIBUTING.md) for development setup and guidelines.
-- Community: [Discord](https://www.gibsonai.com/discord)
+- Community: [Discord](https://discord.gg/abD4eGym6v)
+
+## ⭐️ Star us on GitHub to support the project 
+
+[![Star History Chart](https://api.star-history.com/svg?repos=GibsonAI/memori&type=date&legend=top-left)](https://www.star-history.com/#GibsonAI/memori&type=date&legend=top-left)
 
 ## 📄 License
 
-MIT License - see [LICENSE](./LICENSE) for details.
+Apache 2.0 License - see [LICENSE](./LICENSE) for details.
 
 ---
 

{memorisdk-2.1.1 → memorisdk-2.3.1}/memori/__init__.py

@@ -5,7 +5,7 @@ Professional-grade memory layer with comprehensive error handling, configuration
 management, and modular architecture for production AI systems.
 """
 
-__version__ = "2.1.1"
+__version__ = "2.3.0"
 __author__ = "Harshal More"
 __email__ = "harshalmore2468@gmail.com"
 

{memorisdk-2.1.1 → memorisdk-2.3.1}/memori/agents/conscious_agent.py

@@ -116,7 +116,7 @@ class ConsciouscAgent:
             return False
 
     async def initialize_existing_conscious_memories(
-        self, db_manager, namespace: str = "default"
+        self, db_manager, namespace: str = "default", limit: int = 10
     ) -> bool:
         """
         Initialize by copying ALL existing conscious-info memories to short-term memory
@@ -143,16 +143,17 @@ class ConsciouscAgent:
                 from sqlalchemy import text
 
                 with db_manager._get_connection() as connection:
-                    # Get ALL conscious-info labeled memories from long-term memory
+                    # Get top conscious-info labeled memories from long-term memory (limited for performance)
                     cursor = connection.execute(
                         text(
                             """SELECT memory_id, processed_data, summary, searchable_content,
                                   importance_score, created_at
                            FROM long_term_memory
                            WHERE namespace = :namespace AND classification = 'conscious-info'
-                           ORDER BY importance_score DESC, created_at DESC"""
+                           ORDER BY importance_score DESC, created_at DESC
+                           LIMIT :limit"""
                         ),
-                        {"namespace": namespace},
+                        {"namespace": namespace, "limit": limit},
                     )
                     existing_conscious_memories = cursor.fetchall()
 

{memorisdk-2.1.1 → memorisdk-2.3.1}/memori/agents/memory_agent.py

@@ -204,6 +204,9 @@ CONVERSATION CONTEXT:
                                 "content": f"Process this conversation for enhanced memory storage:\n\n{conversation_text}\n{context_info}",
                             },
                         ],
+                        metadata=[
+                            "INTERNAL_MEMORY_PROCESSING"
+                        ],  # Internal metadata tag
                         response_format=ProcessedLongTermMemory,
                         temperature=0.1,  # Low temperature for consistent processing
                     )
@@ -237,17 +240,19 @@ CONVERSATION CONTEXT:
                 )
 
             logger.debug(
-                f"Processed conversation {chat_id}: "
-                f"classification={processed_memory.classification}, "
-                f"importance={processed_memory.importance}, "
-                f"conscious_context={processed_memory.is_user_context}, "
-                f"promotion_eligible={processed_memory.promotion_eligible}"
+                f"[AGENT] Processed conversation {chat_id[:8]}... - "
+                f"classification: {processed_memory.classification.value} | "
+                f"importance: {processed_memory.importance.value} | "
+                f"conscious_context: {processed_memory.is_user_context} | "
+                f"promotion_eligible: {processed_memory.promotion_eligible}"
             )
 
             return processed_memory
 
         except Exception as e:
-            logger.error(f"Memory agent processing failed for {chat_id}: {e}")
+            logger.error(
+                f"[AGENT] Memory processing failed for {chat_id[:8]}... - {type(e).__name__}: {e}"
+            )
             return self._create_empty_long_term_memory(
                 chat_id, f"Processing failed: {str(e)}"
             )
@@ -307,7 +312,7 @@ CONVERSATION CONTEXT:
 
             if avg_similarity >= similarity_threshold:
                 logger.info(
-                    f"Duplicate detected: {avg_similarity:.2f} similarity with {existing.conversation_id}"
+                    f"[AGENT] Duplicate detected - {avg_similarity:.2f} similarity with {existing.conversation_id[:8]}..."
                 )
                 return existing.conversation_id
 
@@ -415,6 +420,7 @@ CONVERSATION CONTEXT:
                         "content": f"Process this conversation for enhanced memory storage:\n\n{conversation_text}\n{context_info}",
                     },
                 ],
+                metadata=["INTERNAL_MEMORY_PROCESSING"],  # Internal metadata tag
                 temperature=0.1,  # Low temperature for consistent processing
                 max_tokens=2000,  # Ensure enough tokens for full response
             )

{memorisdk-2.1.1 → memorisdk-2.3.1}/memori/agents/retrieval_agent.py

@@ -149,6 +149,9 @@ Be strategic and comprehensive in your search planning."""
                                 "content": prompt,
                             },
                         ],
+                        metadata=[
+                            "INTERNAL_MEMORY_PROCESSING"
+                        ],  # Internal metadata tag
                         response_format=MemorySearchQuery,
                         temperature=0.1,
                     )
@@ -218,15 +221,25 @@ Be strategic and comprehensive in your search planning."""
             all_results = []
             seen_memory_ids = set()
 
-            # For MongoDB and SQL, use the unified search_memories method as primary strategy
-            # This ensures we use the database's native search capabilities
-            logger.debug(f"Executing unified database search using {db_type} manager")
-            primary_results = db_manager.search_memories(
-                query=search_plan.query_text or query, namespace=namespace, limit=limit
-            )
-            logger.debug(
-                f"Primary database search returned {len(primary_results)} results"
-            )
+            # For MongoDB and SQL, use SearchService directly to avoid recursion
+            # This ensures we use the database's native search capabilities without triggering context injection
+            logger.debug(f"Executing direct SearchService search using {db_type}")
+            try:
+                from ..database.search_service import SearchService
+
+                with db_manager.SessionLocal() as session:
+                    search_service = SearchService(session, db_type)
+                    primary_results = search_service.search_memories(
+                        query=search_plan.query_text or query,
+                        namespace=namespace,
+                        limit=limit,
+                    )
+                logger.debug(
+                    f"Direct SearchService returned {len(primary_results)} results"
+                )
+            except Exception as e:
+                logger.error(f"SearchService direct access failed: {e}")
+                primary_results = []
 
             # Process primary results and add search metadata
             for result in primary_results:
@@ -383,9 +396,17 @@ Be strategic and comprehensive in your search planning."""
 
         search_terms = " ".join(keywords)
         try:
-            results = db_manager.search_memories(
-                query=search_terms, namespace=namespace, limit=limit
-            )
+            # Use SearchService directly to avoid recursion
+            from ..database.search_service import SearchService
+
+            db_type = self._detect_database_type(db_manager)
+
+            with db_manager.SessionLocal() as session:
+                search_service = SearchService(session, db_type)
+                results = search_service.search_memories(
+                    query=search_terms, namespace=namespace, limit=limit
+                )
+
             # Ensure results is a list of dictionaries
             if not isinstance(results, list):
                 logger.warning(f"Search returned non-list result: {type(results)}")
@@ -417,14 +438,24 @@ Be strategic and comprehensive in your search planning."""
         if not categories:
             return []
 
-        # This would need to be implemented in the database manager
-        # For now, get all memories and filter by category
+        # Use SearchService directly to avoid recursion
+        # Get all memories and filter by category
         logger.debug(
             f"Searching memories by categories: {categories} in namespace: {namespace}"
         )
-        all_results = db_manager.search_memories(
-            query="", namespace=namespace, limit=limit * 3
-        )
+        try:
+            from ..database.search_service import SearchService
+
+            db_type = self._detect_database_type(db_manager)
+
+            with db_manager.SessionLocal() as session:
+                search_service = SearchService(session, db_type)
+                all_results = search_service.search_memories(
+                    query="", namespace=namespace, limit=limit * 3
+                )
+        except Exception as e:
+            logger.error(f"Category search failed: {e}")
+            all_results = []
 
         logger.debug(
             f"Retrieved {len(all_results)} total results for category filtering"
@@ -628,6 +659,7 @@ Be strategic and comprehensive in your search planning."""
                         "content": prompt,
                     },
                 ],
+                metadata=["INTERNAL_MEMORY_PROCESSING"],  # Internal metadata tag
                 temperature=0.1,
                 max_tokens=1000,  # Ensure enough tokens for full response
             )

{memorisdk-2.1.1 → memorisdk-2.3.1}/memori/core/conversation.py

@@ -207,7 +207,7 @@ class ConversationManager:
             elif mode == "auto":
                 # Auto mode: Search long-term memory database for relevant context
                 logger.debug(
-                    f"Auto-ingest: Processing user input for long-term memory search: '{user_input[:50]}...'"
+                    f"[CONTEXT] Auto-ingest processing - Query: '{user_input[:50]}...' | Session: {session_id[:8]}..."
                 )
                 context = (
                     memori_instance._get_auto_ingest_context(user_input)
@@ -217,11 +217,11 @@ class ConversationManager:
                 if context:
                     context_prompt = self._build_auto_context_prompt(context)
                     logger.debug(
-                        f"Auto-ingest: Successfully injected long-term memory context with {len(context)} items for session {session_id}"
+                        f"[CONTEXT] Long-term memory injected - {len(context)} items | Session: {session_id[:8]}..."
                     )
                 else:
                     logger.debug(
-                        f"Auto-ingest: No relevant memories found in long-term database for query '{user_input[:50]}...' in session {session_id}"
+                        f"[CONTEXT] No relevant memories found for '{user_input[:30]}...' | Session: {session_id[:8]}..."
                     )
 
             # Get conversation history
@@ -242,11 +242,17 @@ class ConversationManager:
                 if previous_messages:
                     system_content += "\n--- Conversation History ---\n"
                     for msg in previous_messages:
-                        role_label = "You" if msg["role"] == "assistant" else "User"
+                        if msg["role"] == "assistant":
+                            role_label = "Assistant"
+                        elif msg["role"] == "user":
+                            role_label = "User"
+                        else:
+                            role_label = msg["role"].capitalize()
                         system_content += f"{role_label}: {msg['content']}\n"
+
                     system_content += "--- End History ---\n"
                     logger.debug(
-                        f"Added {len(previous_messages)} history messages for session {session_id}"
+                        f"[CONTEXT] Added {len(previous_messages)} history messages | Session: {session_id[:8]}..."
                     )
 
             # Find existing system message or create new one
@@ -267,16 +273,17 @@ class ConversationManager:
                     0, {"role": "system", "content": system_content}
                 )
 
+            context_status = "yes" if context_prompt else "no"
+            history_status = "yes" if len(history_messages) > 1 else "no"
             logger.debug(
-                f"Enhanced messages for session {session_id}: context={'yes' if context_prompt else 'no'}, "
-                f"history={'yes' if len(history_messages) > 1 else 'no'}"
+                f"[CONTEXT] Enhanced messages for session {session_id[:8]}... - context: {context_status} | history: {history_status}"
             )
 
             return enhanced_messages
 
         except Exception as e:
             logger.error(
-                f"Failed to inject context with history for session {session_id}: {e}"
+                f"[CONTEXT] Failed to inject context for session {session_id[:8]}... - {type(e).__name__}: {e}"
             )
             return messages