Upload app.py with huggingface_hub

app.py

@@ -1,9 +1,8 @@
 import gradio as gr
-import os, requests, glob
+import os, requests
 from groq import Groq
 
 GROQ_KEY = os.environ.get("GROQ_API_KEY","")
-GITHUB_TOKEN = os.environ.get("GITHUB_TOKEN","")
 client = Groq(api_key=GROQ_KEY)
 
 KNOWHOW = """
@@ -17,19 +16,20 @@ CKD Stages: 1 below 1.5, 2 1.5-3.0, 3-4 3.0-6.0, 5 above 6.0 mg/dL
 Equipment: Heska HT5, time-resolved PIV, Tygon tubing, Arduino
 """
 
-def search_pubmed(query, n=3):
+def search_pubmed(query, n=5):
     try:
         r = requests.get("https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi",
-            params={"db":"pubmed","term":query,"retmax":n,"retmode":"json"}, timeout=10)
+            params={"db":"pubmed","term":query,"retmax":n,"retmode":"json","sort":"date"}, timeout=10)
         ids = r.json()["esearchresult"]["idlist"]
-        if not ids: return ""
+        if not ids: return [], ""
         r2 = requests.get("https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi",
             params={"db":"pubmed","id":",".join(ids),"retmode":"xml","rettype":"abstract"}, timeout=10)
         import xmltodict
         data = xmltodict.parse(r2.content)
         articles = data.get("PubmedArticleSet",{}).get("PubmedArticle",[])
         if isinstance(articles, dict): articles = [articles]
-        out = []
+        real_links = []
+        context = ""
         for a in articles[:n]:
             try:
                 c = a["MedlineCitation"]
@@ -38,66 +38,70 @@ def search_pubmed(query, n=3):
                 if isinstance(abstract, list): abstract = " ".join([str(x) for x in abstract])
                 if isinstance(abstract, dict): abstract = str(abstract.get("#text",""))
                 pmid = str(c["PMID"]["#text"] if isinstance(c["PMID"],dict) else c["PMID"])
-                out.append("[PubMed: "+title[:80]+"]\n"+str(abstract)[:300]+"\nURL: https://pubmed.ncbi.nlm.nih.gov/"+pmid)
+                real_url = "https://pubmed.ncbi.nlm.nih.gov/" + pmid
+                real_links.append("- " + title[:100] + "\n  URL: " + real_url)
+                context += "[PubMed PMID:" + pmid + "] " + title + ". " + str(abstract)[:300] + "\n\n"
             except: continue
-        return "\n\n".join(out)
-    except: return ""
+        return real_links, context
+    except: return [], ""
 
-def search_scholar(query, n=3):
+def search_scholar(query, n=5):
     try:
         r = requests.get("https://api.semanticscholar.org/graph/v1/paper/search",
             params={"query":query,"limit":n,"fields":"title,abstract,year,url"}, timeout=10)
         papers = r.json().get("data",[])
-        out = []
+        real_links = []
+        context = ""
         for p in papers:
-            out.append("[Scholar "+str(p.get("year",""))+": "+p.get("title","")[:80]+"]\n"+(p.get("abstract") or "")[:300]+"\nURL: "+p.get("url",""))
-        return "\n\n".join(out)
-    except: return ""
-
-def search_web_realtime(query):
-    if not GITHUB_TOKEN:
-        return ""
-    try:
-        headers = {
-            "Authorization": "Bearer " + GITHUB_TOKEN,
-            "Content-Type": "application/json"
-        }
-        payload = {
-            "messages": [{"role": "user", "content": query}],
-            "model": "gpt-4o",
-            "tools": [{"type": "bing_search"}]
-        }
-        r = requests.post(
-            "https://api.githubcopilot.com/chat/completions",
-            headers=headers,
-            json=payload,
-            timeout=15
-        )
-        if r.status_code == 200:
-            data = r.json()
-            content = data["choices"][0]["message"]["content"]
-            return "[Real-time Web Search]\n" + str(content)[:600]
-        return ""
-    except Exception as e:
-        return ""
+            title = p.get("title","")
+            abstract = (p.get("abstract") or "")[:300]
+            year = str(p.get("year",""))
+            url = p.get("url","")
+            if url:
+                real_links.append("- " + title[:100] + " (" + year + ")\n  URL: " + url)
+                context += "[Scholar " + year + "] " + title + ". " + abstract + "\n\n"
+        return real_links, context
+    except: return [], ""
 
 def ask_agent(question):
     if not GROQ_KEY:
-        return "Error: GROQ_API_KEY not set."
-    cardio_query = question + " AND (mechanical heart valve OR microfluidic OR CKD creatinine OR PIV hemodynamics OR thrombogenicity)"
-    pubmed = search_pubmed(cardio_query, n=3)
-    scholar = search_scholar(question + " biomedical", n=3)
-    web = search_web_realtime("Latest research on: " + question + " biomedical engineering 2024 2025")
-    all_sources = pubmed + "\n\n" + scholar + "\n\n" + web
+        return "Error: GROQ_API_KEY not set in Space secrets."
+
+    cardio_query = question + " mechanical heart valve OR microfluidic creatinine OR PIV hemodynamics OR thrombogenicity"
+
+    pubmed_links, pubmed_context = search_pubmed(cardio_query, n=5)
+    scholar_links, scholar_context = search_scholar(question + " biomedical", n=5)
+
+    all_context = pubmed_context + scholar_context
+
     response = client.chat.completions.create(
         model="llama-3.3-70b-versatile",
         messages=[
-            {"role":"system","content":"You are CardioLab AI built on Biomni from Stanford SNAP Lab. Expert in SJSU Biomedical Engineering. Always cite sources with URLs when available.\n\nCARDIOLAB KNOW-HOW:\n" + KNOWHOW},
-            {"role":"user","content":"Research question: " + question + "\n\nSources found:\n" + all_sources[:4000]}
+            {"role":"system","content":"""You are CardioLab AI built on Biomni from Stanford SNAP Lab.
+Expert in SJSU Biomedical Engineering research.
+IMPORTANT RULES:
+1. NEVER invent or generate paper titles or URLs
+2. ONLY refer to papers provided in the context below
+3. Always say which source you are using
+4. If you do not know something say so clearly
+
+CARDIOLAB KNOW-HOW:
+""" + KNOWHOW},
+            {"role":"user","content":"Research question: " + question + "\n\nReal papers found (use ONLY these):\n" + all_context[:4000] + "\n\nAnswer the question using only the above sources."}
         ],
-        max_tokens=800
+        max_tokens=600
     )
-    return response.choices[0].message.content
+
+    answer = response.choices[0].message.content
+
+    # Add REAL links section — only verified URLs
+    real_links_section = ""
+    if pubmed_links:
+        real_links_section += "\n\n📚 VERIFIED PUBMED LINKS:\n" + "\n".join(pubmed_links[:5])
+    if scholar_links:
+        real_links_section += "\n\n🎓 VERIFIED SCHOLAR LINKS:\n" + "\n".join(scholar_links[:5])
+
+    return answer + real_links_section
 
 def piv_tool(velocity, shear, hr):
     v = "HIGH - stenosis risk" if float(velocity)>2.0 else "NORMAL"
@@ -116,12 +120,12 @@ def upad_tool(r, g, b):
 
 with gr.Blocks(title="CardioLab AI - SJSU") as demo:
     gr.Markdown("# CardioLab AI Agent")
-    gr.Markdown("### SJSU Biomedical Engineering | Biomni + Llama 70B + PubMed + Web Search")
-    gr.Markdown("GitHub: github.com/pranatechsol/Cardio-Lab-Ai")
+    gr.Markdown("### SJSU Biomedical Engineering | Biomni + Llama 70B + PubMed + Semantic Scholar")
+    gr.Markdown("**All paper links are verified real URLs only** | GitHub: github.com/pranatechsol/Cardio-Lab-Ai")
     with gr.Tab("Research Assistant"):
-        gr.Markdown("### Searches CardioLab papers + PubMed + Semantic Scholar + Real-time Web")
-        q = gr.Textbox(label="Research question", placeholder="e.g. Latest methods for MHV thrombogenicity detection 2025")
-        a = gr.Textbox(label="Answer with citations", lines=12)
+        gr.Markdown("### Searches PubMed + Semantic Scholar — only real verified links")
+        q = gr.Textbox(label="Research question", placeholder="e.g. Methods for MHV thrombogenicity detection")
+        a = gr.Textbox(label="Answer with verified citations", lines=14)
         gr.Button("Search & Answer").click(ask_agent, inputs=q, outputs=a)
     with gr.Tab("PIV Analysis"):
         gr.Markdown("### Analyze PIV flow data from Mock Circulatory Loop")