feat(agent): Add long-term memory feature to Friday

package.json

@@ -55,6 +55,7 @@
         "figlet": "^1.7.0",
         "find-process": "^1.4.10",
         "google-protobuf": "^3.21.4",
+        "json5": "^2.2.3",
         "moment": "^2.30.1",
         "opener": "^1.5.2",
         "portfinder": "^1.0.36",

packages/app/friday/args.py

@@ -1,18 +1,34 @@
 # -*- coding: utf-8 -*-
-import json
+import json5
 from argparse import ArgumentParser, Namespace
+from typing import List, Dict, Any
 
 
 def json_type(value: str) -> dict:
     """Parse a JSON string into a dictionary."""
     if not value or value == "":
         return {}
     try:
-        result = json.loads(value)
+        result = json5.loads(value)
         if not isinstance(result, dict):
             raise ValueError("JSON must be an object/dictionary")
         return result
-    except json.JSONDecodeError as e:
+    except ValueError as e:
+        raise ValueError(f"Invalid JSON string: {e}")
+
+
+def json_list_type(value: str) -> List[Dict[str, Any]]:
+    """Parse a JSON string into a list of dictionaries."""
+    if not value or value == "":
+        return []
+    try:
+        result = json5.loads(value)
+        if not isinstance(result, list) or not all(
+            isinstance(item, dict) for item in result
+        ):
+            raise ValueError("JSON must be an array/list of objects")
+        return result
+    except ValueError as e:
         raise ValueError(f"Invalid JSON string: {e}")
 
 
@@ -49,6 +65,58 @@ def get_args() -> Namespace:
         type=bool,
         required=True,
     )
+    parser.add_argument(
+        "--longTermMemory",
+        type=bool,
+        default=False,
+        help="Whether to enable long-term memory and embedding support.",
+    )
+    parser.add_argument(
+        "--embeddingProvider",
+        type=str,
+        choices=["dashscope", "openai", "gemini", "ollama"],
+        help="Embedding provider name, e.g. openai/dashscope/gemini/ollama.",
+    )
+    parser.add_argument(
+        "--embeddingModelName",
+        type=str,
+        help="Embedding model name, e.g. text-embedding-3-small or text-embedding-v1.",
+    )
+    parser.add_argument(
+        "--embeddingApiKey",
+        type=str,
+        help="API key for the embedding provider; falls back to --apiKey if not set.",
+    )
+    parser.add_argument(
+        "--embeddingKwargs",
+        type=json_type,
+        default={},
+        help="A JSON string for extra kwargs passed to the embedding model (e.g. host, dimensions).",
+    )
+    parser.add_argument(
+        "--saveToLocal",
+        type=bool,
+        default=False,
+        help="Whether to save long-term memory to local disk.",
+    )
+    parser.add_argument(
+        "--localStoragePath",
+        type=str,
+        default="",
+        help="Local storage path for long-term memory.",
+    )
+    parser.add_argument(
+        "--vectorStoreProvider",
+        type=str,
+        default="qdrant",
+        help="Vector store provider for long-term memory (e.g. qdrant, chroma, faiss).",
+    )
+    parser.add_argument(
+        "--vectorStoreKwargs",
+        type=json_type,
+        default={},
+        help="A JSON string for extra kwargs passed to the vector store configuration.",
+    )
     parser.add_argument(
         "--clientKwargs",
         type=json_type,
@@ -61,5 +129,11 @@ def get_args() -> Namespace:
         default={},
         help="A JSON string representing a dictionary of keyword arguments to pass to the LLM generate method.",
     )
+    parser.add_argument(
+        "--mcpServers",
+        type=json_list_type,
+        default=[],
+        help="A JSON string representing a list of MCP server configurations.",
+    )
     args = parser.parse_args()
     return args

packages/app/friday/main.py

@@ -8,7 +8,7 @@
 
 import json5
 from agentscope.agent import ReActAgent
-from agentscope.memory import InMemoryMemory
+from agentscope.memory import InMemoryMemory, Mem0LongTermMemory
 from agentscope.message import Msg
 from agentscope.session import JSONSession
 from agentscope.tool import (
@@ -25,7 +25,7 @@
     studio_post_reply_hook,
 )
 from args import get_args
-from model import get_model, get_formatter
+from model import get_model, get_formatter, get_embedding_model
 from tool.utils import (
     view_agentscope_library,
     view_agentscope_readme,
@@ -35,6 +35,9 @@
 from utils.connect import StudioConnect
 from utils.constants import FRIDAY_SESSION_ID
 
+from mcp_manager import connect_mcp_servers, close_mcp_connections
+from mem0.vector_stores.configs import VectorStoreConfig
+
 
 async def main():
     args = get_args()
@@ -88,14 +91,68 @@ async def main():
         view_agentscope_faq, group_name="agentscope_tools"
     )
 
+    # Get MCP servers configuration and connect
+    mcp_servers = args.mcpServers if hasattr(args, 'mcpServers') else []
+    local_mcp_clients = await connect_mcp_servers(mcp_servers, toolkit)
+
     # get model from args
-    model = get_model(args.llmProvider, args.modelName, args.apiKey, args.clientKwargs, args.generateKwargs)
+    model = get_model(
+        args.llmProvider,
+        args.modelName,
+        args.apiKey,
+        args.clientKwargs,
+        args.generateKwargs,
+    )
     formatter = get_formatter(args.llmProvider)
 
-    # Create the ReAct agent
-    agent = ReActAgent(
-        name="Friday",
-        sys_prompt="""You're Friday, a helpful assistant specialized in daily task management and AgentScope framework support.
+    # Initialize long-term memory if enabled
+    long_term_memory = None
+    if args.longTermMemory:
+        # Create non-streaming model for long-term memory operations
+        memory_model = get_model(
+            args.llmProvider,
+            args.modelName,
+            args.apiKey,
+            args.clientKwargs,
+            args.generateKwargs,
+            stream=False,
+        )
+        embedding_model = get_embedding_model(
+            embeddingProvider=args.embeddingProvider,
+            embeddingModelName=args.embeddingModelName,
+            embeddingApiKey=args.embeddingApiKey,
+            embedding_kwargs=args.embeddingKwargs,
+        )
+
+        # Prepare vector store config with embedding dimensions
+        vector_store_config_dict = {}
+
+        # Add Qdrant-specific configuration
+        if args.vectorStoreProvider == "qdrant":
+            vector_store_config_dict["on_disk"] = args.saveToLocal
+            vector_store_config_dict["path"] = args.localStoragePath
+
+        # Add embedding_model_dims for vector stores that need it (e.g., Qdrant)
+        if (dimensions := getattr(embedding_model, 'dimensions', None)) is not None:
+            vector_store_config_dict["embedding_model_dims"] = dimensions
+
+        # Merge vectorStoreKwargs into the config
+        if args.vectorStoreKwargs:
+            vector_store_config_dict.update(args.vectorStoreKwargs)
+
+        long_term_memory = Mem0LongTermMemory(
+            agent_name="Friday",
+            user_name="Studio",
+            model=memory_model,
+            embedding_model=embedding_model,
+            vector_store_config=VectorStoreConfig(
+                provider=args.vectorStoreProvider,
+                config=vector_store_config_dict,
+            ),
+        )
+
+    # Build system prompt
+    sys_prompt = """You're Friday, a helpful assistant specialized in daily task management and AgentScope framework support.
 
 # Core Objectives
 - Help users manage and complete daily tasks efficiently
@@ -125,7 +182,21 @@ async def main():
 - Never guess or make up implementations
 
 # Available Context
-- Current date and time: {current_time}""".format(
+- Current date and time: {current_time}"""
+
+    # Add long-term memory information if enabled
+    if args.longTermMemory:
+        sys_prompt += """
+
+# Long-term Memory
+- You have long-term memory enabled, which allows you to remember information across conversations
+- Use this capability to provide more personalized and context-aware assistance
+- You can reference past interactions and learned preferences to better serve the user"""
+
+    # Create the ReAct agent
+    agent = ReActAgent(
+        name="Friday",
+        sys_prompt=sys_prompt.format(
             current_time=datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
             max_turns=20,
         ),
@@ -135,6 +206,8 @@ async def main():
         memory=InMemoryMemory(),
         max_iters=50,
         enable_meta_tool=True,
+        long_term_memory=long_term_memory,
+        long_term_memory_mode='both'
     )
 
     path_dialog_history = get_local_file_path("")
@@ -164,6 +237,9 @@ async def main():
         session_id=FRIDAY_SESSION_ID,
         friday=agent
     )
+
+    # Close local MCP connections
+    await close_mcp_connections(local_mcp_clients)
 
 if __name__ == '__main__':
     asyncio.run(main())

packages/app/friday/mcp_manager/__init__.py

@@ -0,0 +1,8 @@
+# -*- coding: utf-8 -*-
+"""MCP module for Friday."""
+from .manager import connect_mcp_servers, close_mcp_connections
+
+__all__ = [
+    'connect_mcp_servers',
+    'close_mcp_connections',
+]