adds interface fixes, sidebar settings , oauth fixes , more graphs , the determinator , and more

README.md

@@ -16,7 +16,7 @@ license: mit
 tags:
   - mcp-in-action-track-enterprise
   - mcp-hackathon
-  - drug-repurposing
+  - deep-research
   - biomedical-ai
   - pydantic-ai
   - llamaindex
@@ -41,11 +41,13 @@ tags:
 
 </div>
 
-# DeepCritical
+# The DETERMINATOR
 
 ## About
 
-The [Deep Critical Gradio Hackathon Team](### Team) met online in the Alzheimer's Critical Literature Review Group in the Hugging Science initiative. We're building the agent framework we want to use for ai assisted research to [turn the vast amounts of clinical data into cures](https://github.com/DeepCritical/GradioDemo).
+The DETERMINATOR is a deep research agent system designed to assist with complex research questions requiring thorough investigation. Originally developed by the Deep Critical Gradio Hackathon Team, The DETERMINATOR specializes in medical research inquiry, functioning as a medical peer junior researcher that helps gather, evaluate, and synthesize evidence from multiple sources.
+
+**Important**: The DETERMINATOR is a research tool and cannot answer medical questions or provide medical advice. It assists researchers by finding and organizing evidence from biomedical literature and clinical trial databases.
 
 For this hackathon we're proposing a simple yet powerful Deep Research Agent that iteratively looks for the answer until it finds it using general purpose websearch and special purpose retrievers for technical retrievers. 
 
@@ -73,7 +75,7 @@ For this hackathon we're proposing a simple yet powerful Deep Research Agent tha
 - [] Apply Deep Research Systems To Generate Short Form Video (up to 5 minutes)
 - [] Visualize Pydantic Graphs as Loading Screens in the UI
 - [] Improve Data Science with more Complex Graph Agents
-- [] Create Deep Critical Drug Reporposing / Discovery Demo
+- [] Create The DETERMINATOR Deep Research Demo
 - [] Create Deep Critical Literal Review
 - [] Create Deep Critical Hypothesis Generator
 - [] Create PyPi Package 

docs/contributing.md

@@ -1,6 +1,6 @@
-# Contributing to DeepCritical
+# Contributing to The DETERMINATOR
 
-Thank you for your interest in contributing to DeepCritical! This guide will help you get started.
+Thank you for your interest in contributing to The DETERMINATOR! This guide will help you get started.
 
 ## Table of Contents
 

docs/getting-started/examples.md

@@ -1,6 +1,6 @@
 # Examples
 
-This page provides examples of using DeepCritical for various research tasks.
+This page provides examples of using The DETERMINATOR for various research tasks.
 
 ## Basic Research Query
 
@@ -11,7 +11,7 @@ This page provides examples of using DeepCritical for various research tasks.
 What are the latest treatments for Alzheimer's disease?
 ```
 
-**What DeepCritical Does**:
+**What The DETERMINATOR Does**:
 1. Searches PubMed for recent papers
 2. Searches ClinicalTrials.gov for active trials
 3. Evaluates evidence quality
@@ -24,7 +24,7 @@ What are the latest treatments for Alzheimer's disease?
 What clinical trials are investigating metformin for cancer prevention?
 ```
 
-**What DeepCritical Does**:
+**What The DETERMINATOR Does**:
 1. Searches ClinicalTrials.gov for relevant trials
 2. Searches PubMed for supporting literature
 3. Provides trial details and status
@@ -40,7 +40,7 @@ Review the evidence for using metformin as an anti-aging intervention,
 including clinical trials, mechanisms of action, and safety profile.
 ```
 
-**What DeepCritical Does**:
+**What The DETERMINATOR Does**:
 1. Uses deep research mode (multi-section)
 2. Searches multiple sources in parallel
 3. Generates sections on:
@@ -56,7 +56,7 @@ including clinical trials, mechanisms of action, and safety profile.
 Test the hypothesis that regular exercise reduces Alzheimer's disease risk.
 ```
 
-**What DeepCritical Does**:
+**What The DETERMINATOR Does**:
 1. Generates testable hypotheses
 2. Searches for supporting/contradicting evidence
 3. Performs statistical analysis (if Modal configured)

docs/getting-started/mcp-integration.md

@@ -1,10 +1,10 @@
 # MCP Integration
 
-DeepCritical exposes a Model Context Protocol (MCP) server, allowing you to use its search tools directly from Claude Desktop or other MCP clients.
+The DETERMINATOR exposes a Model Context Protocol (MCP) server, allowing you to use its search tools directly from Claude Desktop or other MCP clients.
 
 ## What is MCP?
 
-The Model Context Protocol (MCP) is a standard for connecting AI assistants to external tools and data sources. DeepCritical implements an MCP server that exposes its search capabilities as MCP tools.
+The Model Context Protocol (MCP) is a standard for connecting AI assistants to external tools and data sources. The DETERMINATOR implements an MCP server that exposes its search capabilities as MCP tools.
 
 ## MCP Server URL
 
@@ -33,14 +33,14 @@ http://localhost:7860/gradio_api/mcp/
 ~/.config/Claude/claude_desktop_config.json
 ```
 
-### 2. Add DeepCritical Server
+### 2. Add The DETERMINATOR Server
 
 Edit `claude_desktop_config.json` and add:
 
 ```json
 {
   "mcpServers": {
-    "deepcritical": {
+    "determinator": {
       "url": "http://localhost:7860/gradio_api/mcp/"
     }
   }
@@ -53,7 +53,7 @@ Close and restart Claude Desktop for changes to take effect.
 
 ### 4. Verify Connection
 
-In Claude Desktop, you should see DeepCritical tools available:
+In Claude Desktop, you should see The DETERMINATOR tools available:
 - `search_pubmed`
 - `search_clinical_trials`
 - `search_biorxiv`

docs/getting-started/quick-start.md

@@ -1,6 +1,42 @@
-# Quick Start Guide
+# Single Command Deploy
 
-Get up and running with DeepCritical in minutes.
+Deploy with docker instandly with a single command :
+
+```bash
+docker run -it -p 7860:7860 --platform=linux/amd64 \
+    -e DB_KEY="YOUR_VALUE_HERE" \
+    -e SERP_API="YOUR_VALUE_HERE" \
+    -e INFERENCE_API="YOUR_VALUE_HERE" \
+    -e MODAL_TOKEN_ID="YOUR_VALUE_HERE" \
+    -e MODAL_TOKEN_SECRET="YOUR_VALUE_HERE" \
+    -e NCBI_API_KEY="YOUR_VALUE_HERE" \
+    -e SERPER_API_KEY="YOUR_VALUE_HERE" \
+    -e CHROMA_DB_PATH="./chroma_db" \
+    -e CHROMA_DB_HOST="localhost" \
+    -e CHROMA_DB_PORT="8000" \
+    -e RAG_COLLECTION_NAME="deepcritical_evidence" \
+    -e RAG_SIMILARITY_TOP_K="5" \
+    -e RAG_AUTO_INGEST="true" \
+    -e USE_GRAPH_EXECUTION="false" \
+    -e DEFAULT_TOKEN_LIMIT="100000" \
+    -e DEFAULT_TIME_LIMIT_MINUTES="10" \
+    -e DEFAULT_ITERATIONS_LIMIT="10" \
+    -e WEB_SEARCH_PROVIDER="duckduckgo" \
+    -e MAX_ITERATIONS="10" \
+    -e SEARCH_TIMEOUT="30" \
+    -e LOG_LEVEL="DEBUG" \
+    -e EMBEDDING_PROVIDER="local" \
+    -e OPENAI_EMBEDDING_MODEL="text-embedding-3-small" \
+    -e LOCAL_EMBEDDING_MODEL="BAAI/bge-small-en-v1.5" \
+    -e HUGGINGFACE_EMBEDDING_MODEL="sentence-transformers/all-MiniLM-L6-v2" \
+    -e HF_FALLBACK_MODELS="Qwen/Qwen3-Next-80B-A3B-Thinking,Qwen/Qwen3-Next-80B-A3B-Instruct,meta-llama/Llama-3.3-70B-Instruct,meta-llama/Llama-3.1-8B-Instruct,HuggingFaceH4/zephyr-7b-beta,Qwen/Qwen2-7B-Instruct" \
+    -e HUGGINGFACE_MODEL="Qwen/Qwen3-Next-80B-A3B-Thinking" \
+    registry.hf.space/dataquests-deepcritical:latest python src/app.py
+   ```
+
+## Quick start guide
+
+Get up and running with The DETERMINATOR in minutes.
 
 ## Start the Application
 
@@ -99,7 +135,7 @@ What are the active clinical trials investigating Alzheimer's disease treatments
 
 ## Next Steps
 
-- Learn about [MCP Integration](mcp-integration.md) to use DeepCritical from Claude Desktop
+- Learn about [MCP Integration](mcp-integration.md) to use The DETERMINATOR from Claude Desktop
 - Explore [Examples](examples.md) for more use cases
 - Read the [Configuration Guide](../configuration/index.md) for advanced settings
 - Check out the [Architecture Documentation](../architecture/graph-orchestration.md) to understand how it works

docs/index.md

@@ -1,8 +1,10 @@
-# DeepCritical
+# The DETERMINATOR
 
-**AI-Native Drug Repurposing Research Agent**
+**Deep Research Agent for Medical Inquiry**
 
-DeepCritical is a deep research agent system that uses iterative search-and-judge loops to comprehensively answer research questions. The system supports multiple orchestration patterns, graph-based execution, parallel research workflows, and long-running task management with real-time streaming.
+The DETERMINATOR is a deep research agent system that uses iterative search-and-judge loops to comprehensively investigate research questions. The system supports multiple orchestration patterns, graph-based execution, parallel research workflows, and long-running task management with real-time streaming.
+
+**Important**: The DETERMINATOR functions as a medical peer junior researcher that assists with research by gathering and synthesizing evidence. It cannot answer medical questions or provide medical advice.
 
 ## Features
 
@@ -34,7 +36,7 @@ For detailed installation and setup instructions, see the [Getting Started Guide
 
 ## Architecture
 
-DeepCritical uses a Vertical Slice Architecture:
+The DETERMINATOR uses a Vertical Slice Architecture:
 
 1. **Search Slice**: Retrieving evidence from PubMed, ClinicalTrials.gov, and Europe PMC
 2. **Judge Slice**: Evaluating evidence quality using LLMs

docs/overview/architecture.md

@@ -1,6 +1,6 @@
 # Architecture Overview
 
-DeepCritical is a deep research agent system that uses iterative search-and-judge loops to comprehensively answer research questions. The system supports multiple orchestration patterns, graph-based execution, parallel research workflows, and long-running task management with real-time streaming.
+The DETERMINATOR is a deep research agent system that uses iterative search-and-judge loops to comprehensively answer research questions. The system supports multiple orchestration patterns, graph-based execution, parallel research workflows, and long-running task management with real-time streaming.
 
 ## Core Architecture
 

docs/overview/features.md

@@ -1,6 +1,6 @@
 # Features
 
-DeepCritical provides a comprehensive set of features for AI-assisted research:
+The DETERMINATOR provides a comprehensive set of features for AI-assisted research:
 
 ## Core Features
 
@@ -14,7 +14,7 @@ DeepCritical provides a comprehensive set of features for AI-assisted research:
 ### MCP Integration
 
 - **Model Context Protocol**: Expose search tools via MCP server
-- **Claude Desktop**: Use DeepCritical tools directly from Claude Desktop
+- **Claude Desktop**: Use The DETERMINATOR tools directly from Claude Desktop
 - **MCP Clients**: Compatible with any MCP-compatible client
 
 ### Authentication

examples/README.md

@@ -1,8 +1,8 @@
-# DeepCritical Examples
+# The DETERMINATOR Examples
 
 **NO MOCKS. NO FAKE DATA. REAL SCIENCE.**
 
-These demos run the REAL drug repurposing research pipeline with actual API calls.
+These demos run the REAL deep research pipeline with actual API calls.
 
 ---
 
@@ -181,4 +181,4 @@ Mocks belong in `tests/unit/`, not in demos. When you run these examples, you se
 - Real scientific hypotheses
 - Real research reports
 
-This is what DeepCritical actually does. No fake data. No canned responses.
+This is what The DETERMINATOR actually does. No fake data. No canned responses.

examples/full_stack_demo/run_full.py

@@ -1,8 +1,8 @@
 #!/usr/bin/env python3
 """
-Demo: Full Stack DeepCritical Agent (Phases 1-8).
+Demo: Full Stack DETERMINATOR Agent (Phases 1-8).
 
-This script demonstrates the COMPLETE REAL drug repurposing research pipeline:
+This script demonstrates the COMPLETE REAL deep research pipeline:
 - Phase 2: REAL Search (PubMed + ClinicalTrials + Europe PMC)
 - Phase 6: REAL Embeddings (sentence-transformers + ChromaDB)
 - Phase 7: REAL Hypothesis (LLM mechanistic reasoning)

examples/search_demo/run_search.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-Demo: Search for drug repurposing evidence.
+Demo: Search for biomedical research evidence.
 
 This script demonstrates multi-source search functionality:
 - PubMed search (biomedical literature)
@@ -30,7 +30,7 @@ from src.tools.search_handler import SearchHandler
 async def main(query: str) -> None:
     """Run search demo with the given query."""
     print(f"\n{'=' * 60}")
-    print("DeepCritical Search Demo")
+    print("The DETERMINATOR Search Demo")
     print(f"Query: {query}")
     print(f"{'=' * 60}\n")
 
@@ -61,7 +61,7 @@ async def main(query: str) -> None:
 
 if __name__ == "__main__":
     # Default query or use command line arg
-    default_query = "metformin Alzheimer's disease drug repurposing"
+    default_query = "metformin Alzheimer's disease treatment mechanisms"
     query = sys.argv[1] if len(sys.argv) > 1 else default_query
 
     asyncio.run(main(query))

mkdocs.yml

@@ -1,6 +1,6 @@
-site_name: DeepCritical
-site_description: AI-Native Drug Repurposing Research Agent
-site_author: DeepCritical Team
+site_name: The DETERMINATOR
+site_description: Deep Research Agent for Medical Inquiry
+site_author: The DETERMINATOR Team
 site_url: https://deepcritical.github.io/GradioDemo/
 
 repo_name: DeepCritical/GradioDemo

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
-name = "deepcritical"
+name = "determinator"
 version = "0.1.0"
-description = "AI-Native Drug Repurposing Research Agent"
+description = "The DETERMINATOR - Deep Research Agent for Medical Inquiry"
 readme = "README.md"
 requires-python = ">=3.11"
 dependencies = [
@@ -41,6 +41,7 @@ dependencies = [
     "numpy<2.0",
     "llama-index-llms-openai>=0.6.9",
     "llama-index-embeddings-openai>=0.5.1",
+    "ddgs>=9.9.2",
 ]
 
 [project.optional-dependencies]

requirements.txt

@@ -33,8 +33,9 @@ limits>=3.0  # Rate limiting
 pydantic-graph>=1.22.0
 
 # Web search
-duckduckgo-search>=5.0
-
+ddgs>=9.9.2  # duckduckgo-search has been renamed to ddgs
+fake-useragent==2.2.0
+socksio==1.0.0
 # LlamaIndex RAG
 llama-index-llms-huggingface>=0.6.1
 llama-index-llms-huggingface-api>=0.6.1

src/agents/magentic_agents.py

@@ -29,7 +29,7 @@ def create_search_agent(chat_client: Any | None = None) -> ChatAgent:
         name="SearchAgent",
         description=(
             "Searches biomedical databases (PubMed, ClinicalTrials.gov, Europe PMC) "
-            "for drug repurposing evidence"
+            "for research evidence"
         ),
         instructions="""You are a biomedical search specialist. When asked to find evidence:
 
@@ -100,7 +100,7 @@ def create_hypothesis_agent(chat_client: Any | None = None) -> ChatAgent:
 
     return ChatAgent(
         name="HypothesisAgent",
-        description="Generates mechanistic hypotheses for drug repurposing",
+        description="Generates mechanistic hypotheses for research investigation",
         instructions="""You are a biomedical hypothesis generator. Based on evidence:
 
 1. Identify the key molecular targets involved

src/agents/search_agent.py

@@ -28,7 +28,7 @@ class SearchAgent(BaseAgent):  # type: ignore[misc]
     ) -> None:
         super().__init__(
             name="SearchAgent",
-            description="Searches PubMed for drug repurposing evidence",
+            description="Searches PubMed for biomedical research evidence",
         )
         self._handler = search_handler
         self._evidence_store = evidence_store

src/agents/tools.py

@@ -80,7 +80,7 @@ async def search_clinical_trials(query: str, max_results: int = 10) -> str:
     """Search ClinicalTrials.gov for clinical studies.
 
     Use this tool to find ongoing and completed clinical trials
-    for drug repurposing candidates.
+    for research investigation.
 
     Args:
         query: Search terms (e.g., "metformin cancer phase 3")

src/app.py

@@ -1,4 +1,4 @@
-"""Gradio UI for DeepCritical agent with MCP server support."""
+"""Gradio UI for The DETERMINATOR agent with MCP server support."""
 
 import os
 from collections.abc import AsyncGenerator
@@ -737,7 +737,7 @@ def create_demo() -> gr.Blocks:
     Returns:
         Configured Gradio Blocks interface with MCP server and OAuth enabled
     """
-    with gr.Blocks(title="🧬 DeepCritical", fill_height=True) as demo:
+    with gr.Blocks(title="🔬 The DETERMINATOR", fill_height=True) as demo:
         # Add sidebar with login button and information
         # Reference: Working implementation pattern from Gradio docs
         with gr.Sidebar():
@@ -750,96 +750,99 @@ def create_demo() -> gr.Blocks:
             gr.Markdown("---")
             gr.Markdown("### ℹ️ About")  # noqa: RUF001
             gr.Markdown(
-                "AI-Powered Drug Repurposing Agent that searches:\n"
+                "**The DETERMINATOR** - Deep Research Agent for Medical Inquiry\n\n"
+                "Searches:\n"
                 "- PubMed\n"
                 "- ClinicalTrials.gov\n"
-                "- Europe PMC"
+                "- Europe PMC\n\n"
+                "⚠️ **Research tool only** - Cannot answer medical questions or provide medical advice."
             )
-        
-        # Create settings components
-        # Note: ChatInterface doesn't support additional_inputs_accordion parameter in Gradio 6.0
-        # Components are created outside accordion context to ensure they're accessible for additional_inputs
-        mode_radio = gr.Radio(
-            choices=["simple", "advanced", "iterative", "deep", "auto"],
-            value="simple",
-            label="Orchestrator Mode",
-            info=(
-                "Simple: Linear search-judge loop | "
-                "Advanced: Multi-agent (OpenAI) | "
-                "Iterative: Knowledge-gap driven | "
-                "Deep: Parallel sections | "
-                "Auto: Smart routing"
-            ),
-        )
-
-        # Graph mode selection
-        graph_mode_radio = gr.Radio(
-            choices=["iterative", "deep", "auto"],
-            value="auto",
-            label="Graph Research Mode",
-            info="Iterative: Single loop | Deep: Parallel sections | Auto: Detect from query",
-        )
-
-        # Graph execution toggle
-        use_graph_checkbox = gr.Checkbox(
-            value=True,
-            label="Use Graph Execution",
-            info="Enable graph-based workflow execution",
-        )
-
-        # TTS Configuration components
-        # Note: These are created outside accordion to ensure accessibility for additional_inputs
-        # The ChatInterface will display them, but grouping in accordion is not supported via additional_inputs_accordion
-        tts_voice_dropdown = gr.Dropdown(
-            choices=[
-                "af_heart",
-                "af_bella",
-                "af_nicole",
-                "af_aoede",
-                "af_kore",
-                "af_sarah",
-                "af_nova",
-                "af_sky",
-                "af_alloy",
-                "af_jessica",
-                "af_river",
-                "am_michael",
-                "am_fenrir",
-                "am_puck",
-                "am_echo",
-                "am_eric",
-                "am_liam",
-                "am_onyx",
-                "am_santa",
-                "am_adam",
-            ],
-            value=settings.tts_voice,
-            label="TTS Voice",
-            info="Select TTS voice (American English voices: af_*, am_*)",
-            visible=settings.enable_audio_output,
-        )
-        tts_speed_slider = gr.Slider(
-            minimum=0.5,
-            maximum=2.0,
-            value=settings.tts_speed,
-            step=0.1,
-            label="TTS Speech Speed",
-            info="Adjust TTS speech speed (0.5x to 2.0x)",
-            visible=settings.enable_audio_output,
-        )
-        tts_gpu_dropdown = gr.Dropdown(
-            choices=["T4", "A10", "A100", "L4", "L40S"],
-            value=settings.tts_gpu or "T4",
-            label="TTS GPU Type",
-            info="Modal GPU type for TTS (T4 is cheapest, A100 is fastest). Note: GPU changes require app restart.",
-            visible=settings.modal_available and settings.enable_audio_output,
-            interactive=False,  # GPU type set at function definition time, requires restart
-        )
-        enable_audio_output_checkbox = gr.Checkbox(
-            value=settings.enable_audio_output,
-            label="Enable Audio Output",
-            info="Generate audio responses using TTS",
-        )
+            gr.Markdown("---")
+            
+            # Settings Section - Organized in Accordions
+            gr.Markdown("## ⚙️ Settings")
+            
+            # Research Configuration Accordion
+            with gr.Accordion("🔬 Research Configuration", open=True):
+                mode_radio = gr.Radio(
+                    choices=["simple", "advanced", "iterative", "deep", "auto"],
+                    value="simple",
+                    label="Orchestrator Mode",
+                    info=(
+                        "Simple: Linear search-judge loop | "
+                        "Advanced: Multi-agent (OpenAI) | "
+                        "Iterative: Knowledge-gap driven | "
+                        "Deep: Parallel sections | "
+                        "Auto: Smart routing"
+                    ),
+                )
+                
+                graph_mode_radio = gr.Radio(
+                    choices=["iterative", "deep", "auto"],
+                    value="auto",
+                    label="Graph Research Mode",
+                    info="Iterative: Single loop | Deep: Parallel sections | Auto: Detect from query",
+                )
+                
+                use_graph_checkbox = gr.Checkbox(
+                    value=True,
+                    label="Use Graph Execution",
+                    info="Enable graph-based workflow execution",
+                )
+            
+            # Audio/TTS Configuration Accordion
+            with gr.Accordion("🔊 Audio Output", open=False):
+                enable_audio_output_checkbox = gr.Checkbox(
+                    value=settings.enable_audio_output,
+                    label="Enable Audio Output",
+                    info="Generate audio responses using TTS",
+                )
+                
+                tts_voice_dropdown = gr.Dropdown(
+                    choices=[
+                        "af_heart",
+                        "af_bella",
+                        "af_nicole",
+                        "af_aoede",
+                        "af_kore",
+                        "af_sarah",
+                        "af_nova",
+                        "af_sky",
+                        "af_alloy",
+                        "af_jessica",
+                        "af_river",
+                        "am_michael",
+                        "am_fenrir",
+                        "am_puck",
+                        "am_echo",
+                        "am_eric",
+                        "am_liam",
+                        "am_onyx",
+                        "am_santa",
+                        "am_adam",
+                    ],
+                    value=settings.tts_voice,
+                    label="TTS Voice",
+                    info="Select TTS voice (American English voices: af_*, am_*)",
+                )
+                
+                tts_speed_slider = gr.Slider(
+                    minimum=0.5,
+                    maximum=2.0,
+                    value=settings.tts_speed,
+                    step=0.1,
+                    label="TTS Speech Speed",
+                    info="Adjust TTS speech speed (0.5x to 2.0x)",
+                )
+                
+                tts_gpu_dropdown = gr.Dropdown(
+                    choices=["T4", "A10", "A100", "L4", "L40S"],
+                    value=settings.tts_gpu or "T4",
+                    label="TTS GPU Type",
+                    info="Modal GPU type for TTS (T4 is cheapest, A100 is fastest). Note: GPU changes require app restart.",
+                    visible=settings.modal_available,
+                    interactive=False,  # GPU type set at function definition time, requires restart
+                )
 
         # Hidden text components for model/provider (not dropdowns to avoid value mismatch)
         # These will be empty by default and use defaults in configure_orchestrator
@@ -861,6 +864,22 @@ def create_demo() -> gr.Blocks:
             label="🔊 Audio Response",
             visible=settings.enable_audio_output,
         )
+        
+        # Update TTS component visibility based on enable_audio_output_checkbox
+        # This must be after audio_output is defined
+        def update_tts_visibility(enabled: bool) -> tuple[dict[str, Any], dict[str, Any], dict[str, Any]]:
+            """Update visibility of TTS components based on enable checkbox."""
+            return (
+                gr.update(visible=enabled),
+                gr.update(visible=enabled),
+                gr.update(visible=enabled),
+            )
+        
+        enable_audio_output_checkbox.change(
+            fn=update_tts_visibility,
+            inputs=[enable_audio_output_checkbox],
+            outputs=[tts_voice_dropdown, tts_speed_slider, audio_output],
+        )
 
         # Chat interface with multimodal support
         # Examples are provided but will NOT run at startup (cache_examples=False)
@@ -868,12 +887,12 @@ def create_demo() -> gr.Blocks:
         gr.ChatInterface(
             fn=research_agent,
             multimodal=True,  # Enable multimodal input (text + images + audio)
-            title="🧬 DeepCritical",
+            title="🔬 The DETERMINATOR",
             description=(
-                "*AI-Powered Drug Repurposing Agent — searches PubMed, "
+                "*Deep Research Agent for Medical Inquiry — searches PubMed, "
                 "ClinicalTrials.gov & Europe PMC*\n\n"
                 "---\n"
-                "*Research tool only — not for medical advice.*  \n"
+                "*Functions as a medical peer junior researcher. Research tool only — cannot answer medical questions or provide medical advice.*  \n"
                 "**MCP Server Active**: Connect Claude Desktop to `/gradio_api/mcp/`\n\n"
                 "**🎤 Multimodal Support**: Upload images (OCR), record audio (STT), or type text.\n\n"
                 "**⚠️ Authentication Required**: Please **sign in with HuggingFace** above before using this application."
@@ -885,7 +904,7 @@ def create_demo() -> gr.Blocks:
                 # Note: Provider is optional - if empty, HF will auto-select
                 # These examples will NOT run at startup - users must click them after logging in
                 [
-                    "What drugs could be repurposed for Alzheimer's disease?",
+                    "What are the latest research findings on Alzheimer's disease treatments?",
                     "simple",
                     "Qwen/Qwen3-Next-80B-A3B-Thinking",
                     "",

src/legacy_orchestrator.py

@@ -374,7 +374,7 @@ class Orchestrator:
             ]
         )
 
-        return f"""## Drug Repurposing Analysis
+        return f"""## Research Analysis
 
 ### Question
 {query}

src/mcp_tools.py

@@ -1,4 +1,4 @@
-"""MCP tool wrappers for DeepCritical search tools.
+"""MCP tool wrappers for The DETERMINATOR search tools.
 
 These functions expose our search tools via MCP protocol.
 Each function follows the MCP tool contract:
@@ -24,7 +24,7 @@ async def search_pubmed(query: str, max_results: int = 10) -> str:
     Returns titles, authors, abstracts, and citation information.
 
     Args:
-        query: Search query (e.g., "metformin alzheimer", "drug repurposing cancer")
+        query: Search query (e.g., "metformin alzheimer", "cancer treatment mechanisms")
         max_results: Maximum results to return (1-50, default 10)
 
     Returns:
@@ -113,7 +113,7 @@ async def search_all_sources(query: str, max_per_source: int = 5) -> str:
     """Search all biomedical sources simultaneously.
 
     Performs parallel search across PubMed, ClinicalTrials.gov, and Europe PMC.
-    This is the most comprehensive search option for drug repurposing research.
+    This is the most comprehensive search option for deep medical research inquiry.
 
     Args:
         query: Search query (e.g., "metformin alzheimer", "aspirin cancer prevention")
@@ -161,10 +161,10 @@ async def analyze_hypothesis(
     condition: str,
     evidence_summary: str,
 ) -> str:
-    """Perform statistical analysis of drug repurposing hypothesis using Modal.
+    """Perform statistical analysis of research hypothesis using Modal.
 
     Executes AI-generated Python code in a secure Modal sandbox to analyze
-    the statistical evidence for a drug repurposing hypothesis.
+    the statistical evidence for a research hypothesis.
 
     Args:
         drug: The drug being evaluated (e.g., "metformin")

src/orchestrator/graph_orchestrator.py

@@ -705,8 +705,39 @@ class GraphOrchestrator:
         if node.input_transformer:
             input_data = node.input_transformer(input_data)
 
-        # Execute agent
-        result = await node.agent.run(input_data)
+        # Execute agent with error handling
+        try:
+            result = await node.agent.run(input_data)
+        except Exception as e:
+            # Handle validation errors and API errors for planner node
+            if node.node_id == "planner":
+                self.logger.error(
+                    "Planner agent execution failed, using fallback plan",
+                    error=str(e),
+                    error_type=type(e).__name__,
+                )
+                # Return a minimal fallback ReportPlan
+                from src.utils.models import ReportPlan, ReportPlanSection
+                
+                # Extract query from input_data if possible
+                fallback_query = query
+                if isinstance(input_data, str):
+                    # Try to extract query from input string
+                    if "QUERY:" in input_data:
+                        fallback_query = input_data.split("QUERY:")[-1].strip()
+                
+                return ReportPlan(
+                    background_context="",
+                    report_outline=[
+                        ReportPlanSection(
+                            title="Research Findings",
+                            key_question=fallback_query,
+                        )
+                    ],
+                    report_title=f"Research Report: {fallback_query[:50]}",
+                )
+            # For other nodes, re-raise the exception
+            raise
 
         # Transform output if needed
         output = result.output
@@ -855,10 +886,64 @@ class GraphOrchestrator:
             Next node ID
         """
         # Get previous result for decision
-        prev_result = context.get_node_result(context.current_node)
+        # The decision node needs the result from the node that connects to it
+        # Find the previous node by searching edges
+        prev_node_id: str | None = None
+        if self._graph:
+            # Find which node connects to this decision node
+            for from_node, edge_list in self._graph.edges.items():
+                for edge in edge_list:
+                    if edge.to_node == node.node_id:
+                        prev_node_id = from_node
+                        break
+                if prev_node_id:
+                    break
+
+        # Fallback: For continue_decision, it always comes from knowledge_gap
+        if not prev_node_id and node.node_id == "continue_decision":
+            prev_node_id = "knowledge_gap"
+
+        # Get result from previous node (or current node if no previous found)
+        if prev_node_id:
+            prev_result = context.get_node_result(prev_node_id)
+        else:
+            # Fallback: try to get from visited nodes (last visited before current)
+            visited_list = list(context.visited_nodes)
+            if len(visited_list) > 0:
+                prev_node_id = visited_list[-1]
+                prev_result = context.get_node_result(prev_node_id)
+            else:
+                prev_result = context.get_node_result(context.current_node)
+
+        # Handle case where result might be a tuple (from pydantic-graph)
+        # Extract the actual result object if it's a tuple
+        if isinstance(prev_result, tuple) and len(prev_result) > 0:
+            # Check if first element is a KnowledgeGapOutput-like object
+            if hasattr(prev_result[0], "research_complete"):
+                prev_result = prev_result[0]
+            elif len(prev_result) > 1 and hasattr(prev_result[1], "research_complete"):
+                prev_result = prev_result[1]
+            else:
+                # If tuple doesn't contain the object, log warning and use first element
+                self.logger.warning(
+                    "Decision node received tuple result, extracting first element",
+                    node_id=node.node_id,
+                    tuple_length=len(prev_result),
+                )
+                prev_result = prev_result[0]
 
         # Make decision
-        next_node_id = node.decision_function(prev_result)
+        try:
+            next_node_id = node.decision_function(prev_result)
+        except Exception as e:
+            self.logger.error(
+                "Decision function failed",
+                node_id=node.node_id,
+                error=str(e),
+                prev_result_type=type(prev_result).__name__,
+            )
+            # Default to first option on error
+            next_node_id = node.options[0]
 
         # Validate decision
         if next_node_id not in node.options:

src/orchestrator_magentic.py

@@ -122,7 +122,7 @@ class MagenticOrchestrator:
 
         workflow = self._build_workflow()
 
-        task = f"""Research drug repurposing opportunities for: {query}
+        task = f"""Research opportunities for: {query}
 
 Workflow:
 1. SearchAgent: Find evidence from PubMed, ClinicalTrials.gov, and Europe PMC

src/prompts/hypothesis.py

@@ -8,9 +8,11 @@ if TYPE_CHECKING:
     from src.services.embeddings import EmbeddingService
     from src.utils.models import Evidence
 
-SYSTEM_PROMPT = """You are a biomedical research scientist specializing in drug repurposing.
+SYSTEM_PROMPT = """You are a bioinformatics research scientist functioning as a medical peer junior researcher.
 
-Your role is to generate mechanistic hypotheses based on evidence.
+Your role is to generate mechanistic hypotheses and research questions based on evidence.
+
+IMPORTANT: You are a research assistant. You cannot answer medical questions or provide medical advice. Your hypotheses are for research investigation purposes only.
 
 A good hypothesis:
 1. Proposes a MECHANISM: Drug -> Target -> Pathway -> Effect

src/prompts/judge.py

@@ -2,10 +2,11 @@
 
 from src.utils.models import Evidence
 
-SYSTEM_PROMPT = """You are an expert drug repurposing research judge.
+SYSTEM_PROMPT = """You are a medical research evidence evaluator functioning as a peer junior researcher.
 
-Your task is to evaluate evidence from biomedical literature and determine if it's sufficient to
-recommend drug candidates for a given condition.
+Your task is to evaluate evidence from biomedical literature and determine if sufficient evidence has been gathered to synthesize findings for a given research question.
+
+IMPORTANT: You are a research assistant. You cannot answer medical questions or provide medical advice. Your role is to assess whether enough evidence has been collected to support research conclusions.
 
 ## Evaluation Criteria
 
@@ -70,7 +71,7 @@ def format_user_prompt(question: str, evidence: list[Evidence]) -> str:
 
 ## Your Task
 
-Evaluate this evidence and determine if it's sufficient to recommend drug repurposing candidates.
+Evaluate this evidence and determine if it's sufficient to synthesize research findings. Consider the quality, quantity, and relevance of the evidence collected.
 Respond with a JSON object matching the JudgeAssessment schema.
 """
 

src/prompts/report.py

@@ -8,9 +8,11 @@ if TYPE_CHECKING:
     from src.services.embeddings import EmbeddingService
     from src.utils.models import Evidence, MechanismHypothesis
 
-SYSTEM_PROMPT = """You are a scientific writer specializing in drug repurposing research reports.
+SYSTEM_PROMPT = """You are a scientific writer functioning as a medical peer junior researcher, specializing in research report synthesis.
 
-Your role is to synthesize evidence and hypotheses into a clear, structured report.
+Your role is to synthesize evidence and findings into a clear, structured research report.
+
+IMPORTANT: You are a research assistant. You cannot answer medical questions or provide medical advice. Your reports synthesize evidence for research purposes only.
 
 A good report:
 1. Has a clear EXECUTIVE SUMMARY (one paragraph, key takeaways)

src/services/__init__.py

@@ -1 +1 @@
-"""Services for DeepCritical."""
+"""Services for The DETERMINATOR."""

src/services/stt_gradio.py

@@ -23,13 +23,13 @@ class STTService:
         """Initialize STT service.
 
         Args:
-            api_url: Gradio Space URL (default: settings.stt_api_url)
+            api_url: Gradio Space URL (default: settings.stt_api_url or nvidia/canary-1b-v2)
             hf_token: HuggingFace token for authenticated Spaces (default: None)
 
         Raises:
             ConfigurationError: If API URL not configured
         """
-        self.api_url = api_url or settings.stt_api_url
+        self.api_url = api_url or settings.stt_api_url or "https://nvidia-canary-1b-v2.hf.space"
         if not self.api_url:
             raise ConfigurationError("STT API URL not configured")
         self.hf_token = hf_token

src/tools/clinicaltrials.py

@@ -75,7 +75,7 @@ class ClinicalTrialsTool:
                 requests.get,
                 self.BASE_URL,
                 params=params,
-                headers={"User-Agent": "DeepCritical-Research-Agent/1.0"},
+                headers={"User-Agent": "DETERMINATOR-Research-Agent/1.0"},
                 timeout=30,
             )
             response.raise_for_status()

src/tools/search_handler.py

@@ -108,6 +108,17 @@ class SearchHandler:
         sources_searched: list[SourceName] = []
         errors: list[str] = []
 
+        # Map tool names to SourceName values
+        # Some tools have internal names that differ from SourceName literals
+        tool_name_to_source: dict[str, SourceName] = {
+            "duckduckgo": "web",
+            "pubmed": "pubmed",
+            "clinicaltrials": "clinicaltrials",
+            "europepmc": "europepmc",
+            "rag": "rag",
+            "web": "web",  # In case tool already uses "web"
+        }
+
         for tool, result in zip(self.tools, results, strict=True):
             if isinstance(result, Exception):
                 errors.append(f"{tool.name}: {result!s}")
@@ -117,8 +128,14 @@ class SearchHandler:
                 success_result = cast(list[Evidence], result)
                 all_evidence.extend(success_result)
 
-                # Cast tool.name to SourceName (centralized type from models)
-                tool_name = cast(SourceName, tool.name)
+                # Map tool.name to SourceName (handle tool names that don't match SourceName literals)
+                tool_name = tool_name_to_source.get(tool.name, cast(SourceName, tool.name))
+                if tool_name not in ["pubmed", "clinicaltrials", "biorxiv", "europepmc", "preprint", "rag", "web"]:
+                    logger.warning(
+                        "Tool name not in SourceName literals, defaulting to 'web'",
+                        tool_name=tool.name,
+                    )
+                    tool_name = "web"
                 sources_searched.append(tool_name)
                 logger.info("Search tool succeeded", tool=tool.name, count=len(success_result))
 

src/tools/web_search.py

@@ -3,7 +3,11 @@
 import asyncio
 
 import structlog
-from duckduckgo_search import DDGS
+try:
+    from ddgs import DDGS  # New package name
+except ImportError:
+    # Fallback to old package name for backward compatibility
+    from duckduckgo_search import DDGS  # type: ignore[no-redef]
 
 from src.tools.query_utils import preprocess_query
 from src.utils.exceptions import SearchError

src/utils/config.py

@@ -164,6 +164,20 @@ class Settings(BaseSettings):
         description="Modal GPU type for TTS (T4, A10, A100, L4, L40S). None uses default T4.",
     )
 
+    # STT (Speech-to-Text) Configuration
+    stt_api_url: str | None = Field(
+        default="https://nvidia-canary-1b-v2.hf.space",
+        description="Gradio Space URL for STT service (default: nvidia/canary-1b-v2)",
+    )
+    stt_source_lang: str = Field(
+        default="English",
+        description="Source language for STT (full name like 'English', 'Spanish', etc.)",
+    )
+    stt_target_lang: str = Field(
+        default="English",
+        description="Target language for STT (full name like 'English', 'Spanish', etc.)",
+    )
+
     # Report File Output Configuration
     save_reports_to_file: bool = Field(
         default=True,

src/utils/exceptions.py

@@ -1,8 +1,11 @@
-"""Custom exceptions for DeepCritical."""
+"""Custom exceptions for The DETERMINATOR."""
 
 
 class DeepCriticalError(Exception):
-    """Base exception for all DeepCritical errors."""
+    """Base exception for all DETERMINATOR errors.
+
+    Note: Class name kept for backward compatibility.
+    """
 
     pass
 

tests/conftest.py

@@ -43,10 +43,10 @@ def sample_evidence():
             relevance=0.85,
         ),
         Evidence(
-            content="Drug repurposing offers faster path to treatment...",
+            content="Research offers faster path to treatment discovery...",
             citation=Citation(
                 source="pubmed",
-                title="Drug Repurposing Strategies",
+                title="Research Strategies for Treatment Discovery",
                 url="https://example.com/drug-repurposing",
                 date="Unknown",
                 authors=[],

tests/unit/agents/test_report_agent.py

@@ -51,15 +51,15 @@ def sample_hypotheses() -> list[MechanismHypothesis]:
 @pytest.fixture
 def mock_report() -> ResearchReport:
     return ResearchReport(
-        title="Drug Repurposing Analysis: Metformin for Alzheimer's",
+        title="Research Analysis: Metformin for Alzheimer's",
         executive_summary=(
             "This report analyzes metformin as a potential candidate for "
-            "repurposing in Alzheimer's disease treatment. It summarizes "
+            "Alzheimer's disease treatment. It summarizes "
             "findings from mechanistic studies showing AMPK activation effects "
             "and reviews clinical data. The evidence suggests a potential "
             "neuroprotective role, although clinical trials are still limited."
         ),
-        research_question="Can metformin be repurposed for Alzheimer's disease?",
+        research_question="What is the evidence for metformin in Alzheimer's disease treatment?",
         methodology=ReportSection(
             title="Methodology", content="Searched PubMed and web sources..."
         ),

tests/unit/tools/test_rag_tool.py

@@ -46,11 +46,11 @@ class TestRAGTool:
                 },
             },
             {
-                "text": "Drug repurposing offers faster path to treatment.",
+                "text": "Research offers faster path to treatment discovery.",
                 "score": 0.72,
                 "metadata": {
                     "source": "pubmed",
-                    "title": "Drug Repurposing Strategies",
+                    "title": "Research Strategies for Treatment Discovery",
                     "url": "https://example.com/drug-repurposing",
                     "date": "Unknown",
                     "authors": "",

uv.lock

@@ -595,6 +595,26 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f5/10/56978295c14794b2c12007b07f3e41ba26acda9257457d7085b0bb3bb90c/brotli-1.2.0-cp314-cp314-win_amd64.whl", hash = "sha256:e7c0af964e0b4e3412a0ebf341ea26ec767fa0b4cf81abb5e897c9338b5ad6a3", size = 375639, upload-time = "2025-11-05T18:38:55.67Z" },
 ]
 
+[[package]]
+name = "brotlicffi"
+version = "1.2.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cffi" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/84/85/57c314a6b35336efbbdc13e5fc9ae13f6b60a0647cfa7c1221178ac6d8ae/brotlicffi-1.2.0.0.tar.gz", hash = "sha256:34345d8d1f9d534fcac2249e57a4c3c8801a33c9942ff9f8574f67a175e17adb", size = 476682, upload-time = "2025-11-21T18:17:57.334Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e4/df/a72b284d8c7bef0ed5756b41c2eb7d0219a1dd6ac6762f1c7bdbc31ef3af/brotlicffi-1.2.0.0-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:9458d08a7ccde8e3c0afedbf2c70a8263227a68dea5ab13590593f4c0a4fd5f4", size = 432340, upload-time = "2025-11-21T18:17:42.277Z" },
+    { url = "https://files.pythonhosted.org/packages/74/2b/cc55a2d1d6fb4f5d458fba44a3d3f91fb4320aa14145799fd3a996af0686/brotlicffi-1.2.0.0-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:84e3d0020cf1bd8b8131f4a07819edee9f283721566fe044a20ec792ca8fd8b7", size = 1534002, upload-time = "2025-11-21T18:17:43.746Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/9c/d51486bf366fc7d6735f0e46b5b96ca58dc005b250263525a1eea3cd5d21/brotlicffi-1.2.0.0-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:33cfb408d0cff64cd50bef268c0fed397c46fbb53944aa37264148614a62e990", size = 1536547, upload-time = "2025-11-21T18:17:45.729Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/37/293a9a0a7caf17e6e657668bebb92dfe730305999fe8c0e2703b8888789c/brotlicffi-1.2.0.0-cp38-abi3-win32.whl", hash = "sha256:23e5c912fdc6fd37143203820230374d24babd078fc054e18070a647118158f6", size = 343085, upload-time = "2025-11-21T18:17:48.887Z" },
+    { url = "https://files.pythonhosted.org/packages/07/6b/6e92009df3b8b7272f85a0992b306b61c34b7ea1c4776643746e61c380ac/brotlicffi-1.2.0.0-cp38-abi3-win_amd64.whl", hash = "sha256:f139a7cdfe4ae7859513067b736eb44d19fae1186f9e99370092f6915216451b", size = 378586, upload-time = "2025-11-21T18:17:50.531Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/ec/52488a0563f1663e2ccc75834b470650f4b8bcdea3132aef3bf67219c661/brotlicffi-1.2.0.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:fa102a60e50ddbd08de86a63431a722ea216d9bc903b000bf544149cc9b823dc", size = 402002, upload-time = "2025-11-21T18:17:51.76Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/63/d4aea4835fd97da1401d798d9b8ba77227974de565faea402f520b37b10f/brotlicffi-1.2.0.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:7d3c4332fc808a94e8c1035950a10d04b681b03ab585ce897ae2a360d479037c", size = 406447, upload-time = "2025-11-21T18:17:53.614Z" },
+    { url = "https://files.pythonhosted.org/packages/62/4e/5554ecb2615ff035ef8678d4e419549a0f7a28b3f096b272174d656749fb/brotlicffi-1.2.0.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fb4eb5830026b79a93bf503ad32b2c5257315e9ffc49e76b2715cffd07c8e3db", size = 402521, upload-time = "2025-11-21T18:17:54.875Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/d3/b07f8f125ac52bbee5dc00ef0d526f820f67321bf4184f915f17f50a4657/brotlicffi-1.2.0.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:3832c66e00d6d82087f20a972b2fc03e21cd99ef22705225a6f8f418a9158ecc", size = 374730, upload-time = "2025-11-21T18:17:56.334Z" },
+]
+
 [[package]]
 name = "build"
 version = "1.3.0"
@@ -1104,7 +1124,44 @@ wheels = [
 ]
 
 [[package]]
-name = "deepcritical"
+name = "ddgs"
+version = "9.9.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "click" },
+    { name = "fake-useragent" },
+    { name = "httpx", extra = ["brotli", "http2", "socks"] },
+    { name = "lxml" },
+    { name = "primp" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/30/dc/9f83a14164644d3f666b302b25f07909a7ee1307cbd112b147d6ff61b25b/ddgs-9.9.2.tar.gz", hash = "sha256:5b15d2658c68a6ac10ba76d1b870dc413cf6461d3363aa13830eceee900782ba", size = 36017, upload-time = "2025-11-29T13:45:35.644Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e1/0d/708e8cff994138f7e5e901bcaf7ba4063b833ebd4d3b712858434abceb27/ddgs-9.9.2-py3-none-any.whl", hash = "sha256:5fd2bb828a6e3a90bd886109bfdca1b2e62d7932f617e45cda6a5864fcdfcb04", size = 41555, upload-time = "2025-11-29T13:45:34.741Z" },
+]
+
+[[package]]
+name = "defusedxml"
+version = "0.7.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0f/d5/c66da9b79e5bdb124974bfe172b4daf3c984ebd9c2a06e2b8a4dc7331c72/defusedxml-0.7.1.tar.gz", hash = "sha256:1bb3032db185915b62d7c6209c5a8792be6a32ab2fedacc84e01b52c51aa3e69", size = 75520, upload-time = "2021-03-08T10:59:26.269Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/07/6c/aa3f2f849e01cb6a001cd8554a88d4c77c5c1a31c95bdf1cf9301e6d9ef4/defusedxml-0.7.1-py2.py3-none-any.whl", hash = "sha256:a352e7e428770286cc899e2542b6cdaedb2b4953ff269a210103ec58f6198a61", size = 25604, upload-time = "2021-03-08T10:59:24.45Z" },
+]
+
+[[package]]
+name = "deprecated"
+version = "1.2.18"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "wrapt" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/98/97/06afe62762c9a8a86af0cfb7bfdab22a43ad17138b07af5b1a58442690a2/deprecated-1.2.18.tar.gz", hash = "sha256:422b6f6d859da6f2ef57857761bfb392480502a64c3028ca9bbe86085d72115d", size = 2928744, upload-time = "2025-01-27T10:46:25.7Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6e/c6/ac0b6c1e2d138f1002bcf799d330bd6d85084fece321e662a14223794041/Deprecated-1.2.18-py2.py3-none-any.whl", hash = "sha256:bd5011788200372a32418f888e326a09ff80d0214bd961147cfed01b5c018eec", size = 9998, upload-time = "2025-01-27T10:46:09.186Z" },
+]
+
+[[package]]
+name = "determinator"
 version = "0.1.0"
 source = { editable = "." }
 dependencies = [
@@ -1112,6 +1169,7 @@ dependencies = [
     { name = "anthropic" },
     { name = "beautifulsoup4" },
     { name = "chromadb" },
+    { name = "ddgs" },
     { name = "duckduckgo-search" },
     { name = "gradio", extra = ["mcp", "oauth"] },
     { name = "gradio-client" },
@@ -1196,6 +1254,7 @@ requires-dist = [
     { name = "chromadb", specifier = ">=0.4.0" },
     { name = "chromadb", marker = "extra == 'embeddings'", specifier = ">=0.4.0" },
     { name = "chromadb", marker = "extra == 'modal'", specifier = ">=0.4.0" },
+    { name = "ddgs", specifier = ">=9.9.2" },
     { name = "duckduckgo-search", specifier = ">=5.0" },
     { name = "gradio", extras = ["mcp", "oauth"], specifier = ">=6.0.0" },
     { name = "gradio-client", specifier = ">=1.0.0" },
@@ -1259,27 +1318,6 @@ dev = [
     { name = "ty", specifier = ">=0.0.1a28" },
 ]
 
-[[package]]
-name = "defusedxml"
-version = "0.7.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/0f/d5/c66da9b79e5bdb124974bfe172b4daf3c984ebd9c2a06e2b8a4dc7331c72/defusedxml-0.7.1.tar.gz", hash = "sha256:1bb3032db185915b62d7c6209c5a8792be6a32ab2fedacc84e01b52c51aa3e69", size = 75520, upload-time = "2021-03-08T10:59:26.269Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/07/6c/aa3f2f849e01cb6a001cd8554a88d4c77c5c1a31c95bdf1cf9301e6d9ef4/defusedxml-0.7.1-py2.py3-none-any.whl", hash = "sha256:a352e7e428770286cc899e2542b6cdaedb2b4953ff269a210103ec58f6198a61", size = 25604, upload-time = "2021-03-08T10:59:24.45Z" },
-]
-
-[[package]]
-name = "deprecated"
-version = "1.2.18"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "wrapt" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/98/97/06afe62762c9a8a86af0cfb7bfdab22a43ad17138b07af5b1a58442690a2/deprecated-1.2.18.tar.gz", hash = "sha256:422b6f6d859da6f2ef57857761bfb392480502a64c3028ca9bbe86085d72115d", size = 2928744, upload-time = "2025-01-27T10:46:25.7Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/6e/c6/ac0b6c1e2d138f1002bcf799d330bd6d85084fece321e662a14223794041/Deprecated-1.2.18-py2.py3-none-any.whl", hash = "sha256:bd5011788200372a32418f888e326a09ff80d0214bd961147cfed01b5c018eec", size = 9998, upload-time = "2025-01-27T10:46:09.186Z" },
-]
-
 [[package]]
 name = "dirtyjson"
 version = "1.0.8"
@@ -1418,6 +1456,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/c1/ea/53f2148663b321f21b5a606bd5f191517cf40b7072c0497d3c92c4a13b1e/executing-2.2.1-py2.py3-none-any.whl", hash = "sha256:760643d3452b4d777d295bb167ccc74c64a81df23fb5e08eff250c425a4b2017", size = 28317, upload-time = "2025-09-01T09:48:08.5Z" },
 ]
 
+[[package]]
+name = "fake-useragent"
+version = "2.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/41/43/948d10bf42735709edb5ae51e23297d034086f17fc7279fef385a7acb473/fake_useragent-2.2.0.tar.gz", hash = "sha256:4e6ab6571e40cc086d788523cf9e018f618d07f9050f822ff409a4dfe17c16b2", size = 158898, upload-time = "2025-04-14T15:32:19.238Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/51/37/b3ea9cd5558ff4cb51957caca2193981c6b0ff30bd0d2630ac62505d99d0/fake_useragent-2.2.0-py3-none-any.whl", hash = "sha256:67f35ca4d847b0d298187443aaf020413746e56acd985a611908c73dba2daa24", size = 161695, upload-time = "2025-04-14T15:32:17.732Z" },
+]
+
 [[package]]
 name = "fastapi"
 version = "0.122.0"
@@ -2069,6 +2116,18 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/2a/39/e50c7c3a983047577ee07d2a9e53faf5a69493943ec3f6a384bdc792deb2/httpx-0.28.1-py3-none-any.whl", hash = "sha256:d909fcccc110f8c7faf814ca82a9a4d816bc5a6dbfea25d6591d6985b8ba59ad", size = 73517, upload-time = "2024-12-06T15:37:21.509Z" },
 ]
 
+[package.optional-dependencies]
+brotli = [
+    { name = "brotli", marker = "platform_python_implementation == 'CPython'" },
+    { name = "brotlicffi", marker = "platform_python_implementation != 'CPython'" },
+]
+http2 = [
+    { name = "h2" },
+]
+socks = [
+    { name = "socksio" },
+]
+
 [[package]]
 name = "httpx-sse"
 version = "0.4.0"
@@ -5662,6 +5721,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2", size = 10235, upload-time = "2024-02-25T23:20:01.196Z" },
 ]
 
+[[package]]
+name = "socksio"
+version = "1.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f8/5c/48a7d9495be3d1c651198fd99dbb6ce190e2274d0f28b9051307bdec6b85/socksio-1.0.0.tar.gz", hash = "sha256:f88beb3da5b5c38b9890469de67d0cb0f9d494b78b106ca1845f96c10b91c4ac", size = 19055, upload-time = "2020-04-17T15:50:34.664Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/37/c3/6eeb6034408dac0fa653d126c9204ade96b819c936e136c5e8a6897eee9c/socksio-1.0.0-py3-none-any.whl", hash = "sha256:95dc1f15f9b34e8d7b16f06d74b8ccf48f609af32ab33c608d08761c5dcbb1f3", size = 12763, upload-time = "2020-04-17T15:50:31.878Z" },
+]
+
 [[package]]
 name = "soundfile"
 version = "0.13.1"