Upload community_contributions/ederhion_week8/agents/scanner_agent.py with huggingface_hub

community_contributions/ederhion_week8/agents/scanner_agent.py

@@ -0,0 +1,80 @@
+from typing import Optional, List
+from openai import OpenAI
+from agents.deals import ScrapedDeal, DealSelection
+from agents.agent import Agent
+
+class ScannerAgent(Agent):
+    MODEL = "gpt-4o-mini"
+
+    SYSTEM_PROMPT = """You identify and summarize the 5 most detailed used car deals from a list.
+    Respond strictly in JSON (JavaScript Object Notation) format with no explanation. Provide the price as a number derived from the description.
+    Most important is that you respond with the 5 cars that have the most detailed product descriptions (make, model, year, mileage) and a clear price.
+    """
+
+    USER_PROMPT_PREFIX = """Respond with the most promising 5 vehicle deals from this list, selecting those which have detailed descriptions and a price greater than 0.
+    Rephrase the description to be a summary of the vehicle itself.
+    
+    Deals:
+    
+    """
+
+    USER_PROMPT_SUFFIX = "\n\nInclude exactly 5 deals, no more."
+
+    name = "Scanner Agent"
+    color = Agent.CYAN
+
+    def __init__(self):
+        self.log("Scanner Agent is initializing")
+        self.openai = OpenAI()
+        self.log("Scanner Agent is ready")
+
+    def fetch_deals(self, memory) -> List[ScrapedDeal]:
+        self.log("Scanner Agent is about to fetch deals from the automotive RSS feed")
+        urls = [opp.deal.url for opp in memory]
+        scraped = ScrapedDeal.fetch()
+        result = [scrape for scrape in scraped if scrape.url not in urls]
+        self.log(f"Scanner Agent received {len(result)} deals not already scraped")
+        return result
+
+    def make_user_prompt(self, scraped) -> str:
+        user_prompt = self.USER_PROMPT_PREFIX
+        user_prompt += "\n\n".join([scrape.describe() for scraped in scraped])
+        user_prompt += self.USER_PROMPT_SUFFIX
+        return user_prompt
+
+    def scan(self, memory: List[str] = []) -> Optional[DealSelection]:
+        scraped = self.fetch_deals(memory)
+        if scraped:
+            user_prompt = self.make_user_prompt(scraped)
+            self.log("Scanner Agent is calling OpenAI using Structured Outputs")
+            result = self.openai.chat.completions.parse(
+                model=self.MODEL,
+                messages=[
+                    {"role": "system", "content": self.SYSTEM_PROMPT},
+                    {"role": "user", "content": user_prompt},
+                ],
+                response_format=DealSelection,
+                reasoning_effort="none",
+            )
+            parsed_result = result.choices[0].message.parsed
+            parsed_result.deals = [deal for deal in parsed_result.deals if deal.price > 0]
+            self.log(f"Scanner Agent received {len(parsed_result.deals)} selected deals with price>0")
+            return parsed_result
+        return None
+
+    def test_scan(self, memory: List[str] = []) -> Optional[DealSelection]:
+        results = {
+            "deals": [
+                {
+                    "product_description": "2018 Honda Accord EX-L with 45,000 miles. Features a 1.5L turbocharged engine, leather seats, and advanced safety features. Excellent condition.",
+                    "price": 21000,
+                    "url": "https://www.dealnews.com/fake-honda-accord"
+                },
+                {
+                    "product_description": "2020 Toyota Camry SE with 30,000 miles. Includes sport suspension, 18-inch alloy wheels, and a 7-inch touchscreen infotainment system.",
+                    "price": 23500,
+                    "url": "https://www.dealnews.com/fake-toyota-camry"
+                }
+            ]
+        }
+        return DealSelection(**results)