refactor: remove serve decorator name field for simplicity and AgentRegistry

.github/workflows/python.yml

@@ -1,4 +1,4 @@
-name: Python Code Style & Linting
+name: Python Linting & Unit Tests
 
 permissions:
   contents: read
@@ -31,3 +31,27 @@ jobs:
 
     - name: Ruff Lint
       run: ruff check --config ./pyproject.toml .
+
+  unit_test:
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        working-directory: ./python
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version-file: "./python/pyproject.toml"
+
+    - name: Install uv
+      uses: astral-sh/setup-uv@v6
+
+    - name: Install the project
+      run: uv sync --locked --all-extras --dev
+
+    - name: Run tests
+      run: uv run pytest
+

python/valuecell/core/agent/README.md

@@ -18,7 +18,7 @@ The ValueCell Agent System is a distributed intelligent agent framework based on
 ```python
 from valuecell.core.agent.decorator import serve
 
-@serve(name="Calculator Agent", push_notifications=True)
+@serve(push_notifications=True)
 class CalculatorAgent:
     """An agent that can perform basic math calculations"""
 
@@ -84,7 +84,6 @@ The `@serve` decorator is the core tool for creating Agents, providing the follo
 
 ```python
 @serve(
-    name="My Agent",           # Agent name
     host="localhost",          # Service host
     port=9100,                 # Service port (optional, auto-allocated)
     streaming=True,            # Whether to support streaming response
@@ -187,14 +186,14 @@ from valuecell.core.agent.connect import RemoteConnections
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
-@serve(name="Calculator Agent", push_notifications=True)
+@serve(push_notifications=True)
 class CalculatorAgent:
     async def stream(self, query, session_id, task_id):
         yield {"is_task_complete": False, "content": f"🧮 Calculating: {query}"}
         await asyncio.sleep(0.5)
         yield {"is_task_complete": True, "content": "✅ Calculation complete"}
 
-@serve(name="Weather Agent", port=9101, push_notifications=True)
+@serve(port=9101, push_notifications=True)
 class WeatherAgent:
     async def stream(self, query, session_id, task_id):
         yield {"is_task_complete": False, "content": f"🌤️ Checking weather: {query}"}

python/valuecell/core/agent/connect.py

@@ -9,7 +9,7 @@
 from a2a.types import AgentCard
 from valuecell.core.agent.client import AgentClient
 from valuecell.core.agent.listener import NotificationListener
-from valuecell.core.agent.registry import AgentRegistry
+from valuecell.core.agent import registry
 from valuecell.utils import get_agent_card_path, get_next_available_port
 
 logger = logging.getLogger(__name__)
@@ -143,7 +143,7 @@ async def start_agent(
             return await self._handle_remote_agent(agent_name)
 
         # Handle local agent
-        agent_class = AgentRegistry.get_agent(agent_name)
+        agent_class = registry.get_agent_class_by_name(agent_name)
         if not agent_class:
             raise ValueError(f"Agent '{agent_name}' not found in registry")
 
@@ -314,7 +314,7 @@ def list_available_agents(self) -> List[str]:
         if not self._remote_agent_configs:
             self._load_remote_agent_configs()
 
-        local_agents = AgentRegistry.list_agents()
+        local_agents = registry.list_agent_names()
         remote_agents = list(self._remote_agent_configs.keys())
         return local_agents + remote_agents
 
@@ -346,9 +346,11 @@ def get_agent_info(self, agent_name: str) -> dict:
                 "name": agent_name,
                 "type": "remote",
                 "url": config_data.get("url"),
-                "card": agent_card.model_dump(exclude_none=True)
-                if agent_card
-                else config_data,
+                "card": (
+                    agent_card.model_dump(exclude_none=True)
+                    if agent_card
+                    else config_data
+                ),
                 "connected": agent_name in self._connections,
                 "running": False,  # Remote agents are not managed by us
                 "has_listener": False,

python/valuecell/core/agent/decorator.py

@@ -26,7 +26,7 @@
 )
 from a2a.utils import new_agent_text_message, new_task
 from a2a.utils.errors import ServerError
-from valuecell.core.agent.registry import AgentRegistry
+from valuecell.core.agent import registry
 from valuecell.core.agent.types import BaseAgent
 from valuecell.utils import (
     get_agent_card_path,
@@ -38,7 +38,6 @@
 
 
 def serve(
-    name: str = None,
     host: str = "localhost",
     port: int = None,
     streaming: bool = True,
@@ -64,7 +63,7 @@ def decorator(cls: Type) -> Type:
                     agent_skills.append(skill)
 
         # Determine the agent name consistently
-        agent_name = name or cls.__name__
+        agent_name = cls.__name__
 
         # Create decorated class
         class DecoratedAgent(cls):
@@ -132,12 +131,9 @@ async def serve(self):
         DecoratedAgent.__name__ = cls.__name__
         DecoratedAgent.__qualname__ = cls.__qualname__
 
-        # Store agent name as class attribute for registry management
-        DecoratedAgent.__agent_name__ = agent_name
-
         # Register to registry
         try:
-            AgentRegistry.register(DecoratedAgent, agent_name)
+            registry.register(DecoratedAgent, agent_name)
         except ImportError:
             # Registry not available, skip registration
             logger.warning(