feat: new implementation of simple AI agent that can follow a goal and return

experiments/agents/agent.py

@@ -1,26 +1,44 @@
-from base import Agent
+from os import environ
+from base import Agent as BaseAgent
+from browser_use import Browser, Agent, ChatOpenAI
+from enum import Enum
 
+class AgentTypes(str, Enum):
+    GENERIC_BROWSER_USE_AGENT = "generic_browser_use_agent"
 
-class GenericComputerUseAgent(Agent):
-    # TODO: implement code using computer-use library
-    """
-    async def do_job(jtbd):
-    browser = Browser(headless=True)
-    llm = ChatOpenAI(model="gpt-5-mini")
-    task = "You are an agent for doing anything on http://localhost:3000 (wait for the website to load) for the user. The user wants you to do the following job: "+jtbd + ". Use the browser to do it. Be careful to follow all instructions and complete the job fully. You have the power to purchase or to do anything."
-    agent = Agent(task=task, llm=llm, browser=browser)
-    return await agent.run()
+def _build_prompt(goal : str, environment_url : str) -> str:
+    #TODO: Improve prompt engineering here and experiment with
+    return f"""You are an autonomous agent tasked with achieving the following goal: {goal}
+You have access to a web browser to interact with the environment at {environment_url}.
+Use the browser to navigate, gather information, and perform actions necessary to accomplish your goal.
+Be thorough and ensure you complete the task fully."""
+
+class GenericBrowserUseAgent(BaseAgent):
+    def __init__(self,
+                 goal: str,
+                 url: str = "http://localhost:3000",
+                 timeout: int = 300,
+                 llm_model: str = "gpt-5-mini",
+                 headless: bool = True):
+        super().__init__(goal, url, timeout)
+        self.llm_model = ChatOpenAI(model=llm_model)
+        self.browser = Browser(headless=headless)
+        self.agent = Agent(task=_build_prompt(goal, url),
+                           llm=self.llm_model,
+                           browser=self.browser)
+    async def act(self) -> str:
+        self.result = await self.agent.run()
+        return self.result.final_result()
+
+def get_agent(agent_type: AgentTypes,  **kwargs) -> Agent:
+    if agent_type == AgentTypes.GENERIC_BROWSER_USE_AGENT:
+        return GenericBrowserUseAgent(**kwargs)
+    else:
+        raise ValueError(f"Unknown agent type: {agent_type}")
 
 if __name__ == "__main__":
-    JTBD= "Book the best room"
-    R=asyncio.run(do_job(JTBD))
-    print(R.final_result())
-    """
-    pass
-
-
-
-
-def get_agent():
-    # construct
-    pass
+    import asyncio
+    JTBD= "Name of the company of this website"
+    agent = get_agent(AgentTypes.GENERIC_BROWSER_USE_AGENT, goal=JTBD, url="https://ie.edu", timeout=300)
+    R=asyncio.run(agent.act())
+    print(R)

experiments/agents/base.py

@@ -1,27 +1,19 @@
-import asyncio
+from abc import ABC, abstractmethod
+from typing import Optional
 
-class Agent:
-    def __init__(self,
-                 goal : str = "Get Information on All Prices",
-                 environment_url : str = "https://www.example.com",
-                 timeout : int = 60 * 5):
+class Agent(ABC):
+    """Base interface for browser automation agents"""
+
+    def __init__(self, goal: str, url: str = "http://localhost:3000", timeout: int = 300):
         self.goal = goal
-        self.environment_url = environment_url
+        self.url = url
         self.timeout = timeout
-        self.result = None
-        # TODO: implement agent initialization
+        self.result: Optional[str] = None
+
+    @abstractmethod
+    async def act(self) -> str:
+        """Execute goal and return result text"""
         pass
-    async def act(self):
-        # set the self.result to whatever text result the agents returns
-        pass # return await _async_method()
 
-    def final_result(self) -> str|None:
+    def final_result(self) -> Optional[str]:
         return self.result
-
-
-# asyncio.run(Agent(...).act())
-if __name__ == "__main__":
-    print("Testing Agent...")
-    agent = Agent(goal="Find the best price for a laptop", environment_url="https://www.example.com")
-    asyncio.run(agent.act())
-    print(f"Agent Result: {agent.final_result()}")