docs: update demos to use all 3 search sources

- Remove stale DuckDuckGo references from README
- Update all demos to use PubMed + ClinicalTrials + bioRxiv
- Update docstrings and summary messages
- All 76 tests passing

Files updated:
- examples/README.md
- examples/full_stack_demo/run_full.py
- examples/hypothesis_demo/run_hypothesis.py
- examples/orchestrator_demo/run_agent.py
- examples/orchestrator_demo/run_magentic.py

examples/README.md

@@ -28,16 +28,17 @@ NCBI_API_KEY=your-key
 
 ### 1. Search Demo (No LLM Required)
 
-Demonstrates REAL parallel search across PubMed and Web.
+Demonstrates REAL parallel search across PubMed, ClinicalTrials.gov, and bioRxiv/medRxiv.
 
 ```bash
 uv run python examples/search_demo/run_search.py "metformin cancer"
 ```
 
 **What's REAL:**
-- Actual NCBI E-utilities API calls
-- Actual DuckDuckGo web searches
-- Real papers, real URLs, real content
+- Actual NCBI E-utilities API calls (PubMed)
+- Actual ClinicalTrials.gov API calls
+- Actual bioRxiv/medRxiv preprint API calls
+- Real papers, real trials, real preprints
 
 ---
 
@@ -67,7 +68,7 @@ uv run python examples/orchestrator_demo/run_agent.py "aspirin alzheimer" --iter
 ```
 
 **What's REAL:**
-- Real PubMed + Web searches
+- Real PubMed + ClinicalTrials + bioRxiv searches
 - Real LLM judge evaluating evidence quality
 - Real iterative refinement based on LLM decisions
 - Real research synthesis
@@ -117,7 +118,7 @@ uv run python examples/full_stack_demo/run_full.py "sildenafil heart failure" -i
 ```
 
 **What's REAL:**
-1. Real PubMed + Web evidence collection
+1. Real PubMed + ClinicalTrials + bioRxiv evidence collection
 2. Real embedding-based semantic deduplication
 3. Real LLM mechanistic hypothesis generation
 4. Real LLM evidence quality assessment
@@ -146,7 +147,7 @@ Output: Publication-quality research report with validated citations.
 User Query
     |
     v
-[REAL Search] --> Actual PubMed + Web API calls
+[REAL Search] --> PubMed + ClinicalTrials + bioRxiv APIs
     |
     v
 [REAL Embeddings] --> Actual sentence-transformers

examples/full_stack_demo/run_full.py

@@ -3,7 +3,7 @@
 Demo: Full Stack DeepCritical Agent (Phases 1-8).
 
 This script demonstrates the COMPLETE REAL drug repurposing research pipeline:
-- Phase 2: REAL Search (PubMed only)
+- Phase 2: REAL Search (PubMed + ClinicalTrials + bioRxiv)
 - Phase 6: REAL Embeddings (sentence-transformers + ChromaDB)
 - Phase 7: REAL Hypothesis (LLM mechanistic reasoning)
 - Phase 3: REAL Judge (LLM evidence assessment)
@@ -116,13 +116,17 @@ async def run_full_demo(query: str, max_iterations: int) -> None:
     from src.agents.hypothesis_agent import HypothesisAgent
     from src.agents.report_agent import ReportAgent
     from src.services.embeddings import EmbeddingService
+    from src.tools.biorxiv import BioRxivTool
+    from src.tools.clinicaltrials import ClinicalTrialsTool
     from src.tools.pubmed import PubMedTool
     from src.tools.search_handler import SearchHandler
 
     # Initialize REAL services
     print("[Init] Loading embedding model...")
     embedding_service = EmbeddingService()
-    search_handler = SearchHandler(tools=[PubMedTool()], timeout=30.0)
+    search_handler = SearchHandler(
+        tools=[PubMedTool(), ClinicalTrialsTool(), BioRxivTool()], timeout=30.0
+    )
     judge_handler = JudgeHandler()
 
     # Shared evidence store
@@ -133,7 +137,7 @@ async def run_full_demo(query: str, max_iterations: int) -> None:
         print_step(iteration, f"ITERATION {iteration}/{max_iterations}")
 
         # Step 1: REAL Search
-        print("\n[Search] Querying PubMed (REAL API calls)...")
+        print("\n[Search] Querying PubMed + ClinicalTrials + bioRxiv (REAL API calls)...")
         all_evidence = await _run_search_iteration(
             query, iteration, evidence_store, all_evidence, search_handler, embedding_service
         )
@@ -223,7 +227,7 @@ Examples:
     print("  DeepCritical Full Stack Demo Complete!")
     print("  ")
     print("  Everything you just saw was REAL:")
-    print("    - Real PubMed searches")
+    print("    - Real PubMed + ClinicalTrials + bioRxiv searches")
     print("    - Real embedding computations")
     print("    - Real LLM reasoning")
     print("    - Real scientific report")

examples/hypothesis_demo/run_hypothesis.py

@@ -3,7 +3,7 @@
 Demo: Hypothesis Generation (Phase 7).
 
 This script demonstrates the REAL hypothesis generation pipeline:
-1. REAL search: PubMed (actual API calls)
+1. REAL search: PubMed + ClinicalTrials + bioRxiv (actual API calls)
 2. REAL embeddings: Semantic deduplication
 3. REAL LLM: Mechanistic hypothesis generation
 
@@ -21,6 +21,8 @@ from typing import Any
 
 from src.agents.hypothesis_agent import HypothesisAgent
 from src.services.embeddings import EmbeddingService
+from src.tools.biorxiv import BioRxivTool
+from src.tools.clinicaltrials import ClinicalTrialsTool
 from src.tools.pubmed import PubMedTool
 from src.tools.search_handler import SearchHandler
 
@@ -35,8 +37,10 @@ async def run_hypothesis_demo(query: str) -> None:
         print(f"{'='*60}\n")
 
         # Step 1: REAL Search
-        print("[Step 1] Searching PubMed...")
-        search_handler = SearchHandler(tools=[PubMedTool()], timeout=30.0)
+        print("[Step 1] Searching PubMed + ClinicalTrials + bioRxiv...")
+        search_handler = SearchHandler(
+            tools=[PubMedTool(), ClinicalTrialsTool(), BioRxivTool()], timeout=30.0
+        )
         result = await search_handler.execute(query, max_results_per_tool=5)
 
         print(f"  Found {result.total_found} results from {result.sources_searched}")
@@ -128,7 +132,7 @@ Examples:
 
     print("\n" + "=" * 60)
     print("Demo complete! This was a REAL pipeline:")
-    print("  1. REAL search: Actual PubMed API calls")
+    print("  1. REAL search: PubMed + ClinicalTrials + bioRxiv APIs")
     print("  2. REAL embeddings: Actual sentence-transformers")
     print("  3. REAL LLM: Actual hypothesis generation")
     print("=" * 60 + "\n")

examples/orchestrator_demo/run_agent.py

@@ -3,7 +3,7 @@
 Demo: DeepCritical Agent Loop (Search + Judge + Orchestrator).
 
 This script demonstrates the REAL Phase 4 orchestration:
-- REAL Iterative Search (PubMed only)
+- REAL Iterative Search (PubMed + ClinicalTrials + bioRxiv)
 - REAL Evidence Evaluation (LLM Judge)
 - REAL Orchestration Loop
 - REAL Final Synthesis
@@ -24,6 +24,8 @@ import sys
 
 from src.agent_factory.judges import JudgeHandler
 from src.orchestrator import Orchestrator
+from src.tools.biorxiv import BioRxivTool
+from src.tools.clinicaltrials import ClinicalTrialsTool
 from src.tools.pubmed import PubMedTool
 from src.tools.search_handler import SearchHandler
 from src.utils.models import OrchestratorConfig
@@ -38,7 +40,7 @@ async def main() -> None:
         formatter_class=argparse.RawDescriptionHelpFormatter,
         epilog="""
 This demo runs the REAL search-judge-synthesize loop:
-  1. REAL search: Actual PubMed queries
+  1. REAL search: PubMed + ClinicalTrials + bioRxiv queries
   2. REAL judge: Actual LLM assessing evidence quality
   3. REAL loop: Actual iterative refinement based on LLM decisions
   4. REAL synthesis: Actual research summary generation
@@ -77,7 +79,9 @@ Examples:
     print(f"{'='*60}\n")
 
     # Setup REAL components
-    search_handler = SearchHandler(tools=[PubMedTool()], timeout=30.0)
+    search_handler = SearchHandler(
+        tools=[PubMedTool(), ClinicalTrialsTool(), BioRxivTool()], timeout=30.0
+    )
     judge_handler = JudgeHandler()  # REAL LLM judge
 
     config = OrchestratorConfig(max_iterations=args.iterations)
@@ -101,7 +105,7 @@ Examples:
 
     print("\n" + "=" * 60)
     print("Demo complete! Everything was REAL:")
-    print("  - Real PubMed searches")
+    print("  - Real PubMed + ClinicalTrials + bioRxiv searches")
     print("  - Real LLM judge decisions")
     print("  - Real iterative refinement")
     print("=" * 60 + "\n")

examples/orchestrator_demo/run_magentic.py

@@ -18,6 +18,8 @@ import sys
 
 from src.agent_factory.judges import JudgeHandler
 from src.orchestrator_factory import create_orchestrator
+from src.tools.biorxiv import BioRxivTool
+from src.tools.clinicaltrials import ClinicalTrialsTool
 from src.tools.pubmed import PubMedTool
 from src.tools.search_handler import SearchHandler
 from src.utils.models import OrchestratorConfig
@@ -42,7 +44,9 @@ async def main() -> None:
     print(f"{ '='*60}\n")
 
     # 1. Setup Search Tools
-    search_handler = SearchHandler(tools=[PubMedTool()], timeout=30.0)
+    search_handler = SearchHandler(
+        tools=[PubMedTool(), ClinicalTrialsTool(), BioRxivTool()], timeout=30.0
+    )
 
     # 2. Setup Judge
     judge_handler = JudgeHandler()