update hybrid_retriever_tool file

tools/hybrid_retriever_tool.py

@@ -6,6 +6,7 @@ from openai import OpenAI
 from crewai_tools import RagTool
 from pydantic import Field, PrivateAttr
 import os
+import re
 
 class HybridRetrieverTool(RagTool):
     name: str = "Hybrid Retriever Tool"
@@ -60,12 +61,28 @@ class HybridRetrieverTool(RagTool):
     def summarize_passages(self, topic: str, passages):
         if isinstance(passages, str):
             passages = [passages]
-        text_block = "\n".join(passages)
+        # 🧹 Clean each passage (remove links, HTML tags, redundant whitespace)
+        clean_passages = []
+        for p in passages:
+            p = re.sub(r"http\S+", "", p)                  # remove URLs
+            p = re.sub(r"\s+", " ", p).strip()             # normalize spaces
+            p = re.sub(r"[^A-Za-z0-9.,!?;:()\-\s]", "", p) # strip stray symbols
+            clean_passages.append(p)
+        # Build condensed input (limit total tokens)
+        text_block = " ".join(clean_passages[:5])[:4000]
         try:
             response = self._client.chat.completions.create(
                 model="gpt-4o-mini",
                 messages=[
-                    {"role": "system", "content": "You are an expert summarizer."},
+                    {
+                        "role": "system",
+                        "content": (
+                            "You are a concise research summarizer. "
+                            "Produce a 1–2 paragraph overview that highlights key facts, "
+                            "themes, and findings relevant to the topic. "
+                            "Exclude URLs, lists, HTML remnants, or boilerplate text."
+                        ),
+                    },
                     {"role": "user", "content": f"Summarize these passages about {topic}:\n\n{text_block}"}
                 ],
                 temperature=0.3