feat(agent): add web search agent with LangGraph and Tavily integration

- Created agent module with websearch_agent and run() entrypoint
- Implemented Tavily search tool (max_results=10)
- Updated Gradio app to use the new agent instead of fixed answers
- Added dependencies: langchain, langchain-google-genai, langgraph, tavily-python
- Added .env.example for API key configuration (GOOGLE_API_KEY, TAVILY_API_KEY)
- Updated README with setup and usage documentation
- Removed old template files (main.py, pyproject.toml)

.env.example

@@ -0,0 +1,2 @@
+GOOGLE_API_KEY=your-google-api-key-here
+TAVILY_API_KEY=your-tavily-api-key-here

.gitignore

@@ -0,0 +1 @@
+.env

README.md

@@ -15,9 +15,64 @@ hf_oauth_expiration_minutes: 480
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
-## Start in local
+## Setup
+
+1. Install dependencies:
 
 ```bash
 uv pip install -r requirements.txt
+```
+
+2. Copy `.env.example` to `.env` and fill in your API keys:
+
+```bash
+cp .env.example .env
+```
+
+```
+GOOGLE_API_KEY=your-google-api-key-here
+TAVILY_API_KEY=your-tavily-api-key-here
+```
+
+## Usage
+
+### Run Gradio app locally
+
+```bash
 uv run app.py
+```
+
+### Run the agent directly
+
+```bash
+uv run python -m agent.agent
+```
+
+### Test the search tool only
+
+```bash
+uv run python -m agent.tools.search
+```
+
+### Use in code
+
+```python
+from agent import run
+
+answer = run("What is LangGraph?")
+print(answer)
+```
+
+## Project Structure
+
+```
+agent/
+├── __init__.py          # Exports run()
+├── agent.py             # websearch_agent + run entrypoint
+└── tools/
+    ├── __init__.py      # Exports web_search
+    └── search.py        # Tavily search tool (max_results=10)
+app.py                   # Gradio UI for HuggingFace Spaces
+.env.example             # API keys template
+requirements.txt         # Python dependencies
 ```

agent/__init__.py

@@ -0,0 +1,3 @@
+from agent.agent import run
+
+__all__ = ["run"]

agent/agent.py

@@ -0,0 +1,40 @@
+import os
+from dotenv import load_dotenv
+from langchain.agents import create_agent
+from agent.tools.search import web_search
+
+load_dotenv()
+
+
+def websearch_agent(query: str) -> str:
+    """Create a simple agent with Gemini and Tavily search."""
+    base_agent = create_agent(
+        model="google_genai:gemini-3-flash-preview",
+        tools=[web_search],
+    )
+    result = base_agent.invoke(
+        {
+            "messages": [
+                {
+                    "role": "system",
+                    "content": "Try to search web sites and get the answer.",
+                },
+                {"role": "user", "content": query},
+            ]
+        }
+    )
+    content = result["messages"][-1].content[0]
+    return content["text"]
+
+
+def run(query: str) -> str:
+    """Let Multiple agents finish the work"""
+    result = ""
+    result = websearch_agent(query)
+    return result
+
+
+if __name__ == "__main__":
+    query = input("Query:\n")
+    answer = run(query)
+    print(answer)

agent/tools/__init__.py

@@ -0,0 +1,3 @@
+from agent.tools.search import web_search
+
+__all__ = ["web_search"]

agent/tools/search.py

@@ -0,0 +1,74 @@
+import os
+from langchain_core.tools import tool
+from tavily import TavilyClient  # type: ignore[import]
+
+
+@tool
+def web_search(query: str, depth: str = "advanced", max_results: int = 5) -> str:
+    """
+    Search the web using Tavily and extract full Markdown content from top results.
+    Useful for in-depth analysis, table data, or detailed technical documentation.
+
+    Args:
+        query: The search query string.
+        depth: Search depth, either "basic" or "advanced".
+        max_results: Number of results to return (recommended 3-5 to save tokens).
+    """
+    print(f"[Search & Extract] {query}")
+    client = TavilyClient(api_key=os.environ["TAVILY_API_KEY"])
+
+    # 1. Search and get URLs
+    search_response = client.search(
+        query=query, search_depth=depth, max_results=max_results
+    )
+
+    results = search_response.get("results", [])
+    if not results:
+        return "No results found."
+
+    urls = [r["url"] for r in results]
+
+    # 2. Extract full content via Extract API (max 20 URLs per call)
+    try:
+        extraction = client.extract(
+            urls=urls,
+            extract_depth="advanced",  # For tables and structured data
+            format="markdown",  # Most readable format for LLMs
+        )
+
+        extracted_results = {
+            item["url"]: item["raw_content"] for item in extraction.get("results", [])
+        }
+    except Exception as e:
+        print(f"Extraction failed: {e}")
+        extracted_results = {}
+
+    # 3. Format output
+    final_output = []
+    for r in results:
+        url = r["url"]
+        title = r["title"]
+        snippet = r["content"]  # Original search snippet
+        full_content = extracted_results.get(url, "Full content extraction failed.")
+
+        content_block = (
+            f"### Title: {title}\n"
+            f"**URL:** {url}\n"
+            f"**Snippet:** {snippet}\n\n"
+            f"**Full Markdown Content:**\n\n{full_content}\n"
+            f"{'=' * 50}"
+        )
+        final_output.append(content_block)
+
+    return "\n\n".join(final_output)
+
+
+if __name__ == "__main__":
+    from dotenv import load_dotenv
+
+    load_dotenv()
+
+    # Test run
+    test_query = "What is LangGraph?"
+    result = web_search.invoke({"query": test_query})
+    print(result)

app.py

@@ -8,16 +8,17 @@ import pandas as pd
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
+# --- Agent Definition ---
+from agent.agent import run as agent_run
+
 class BasicAgent:
     def __init__(self):
         print("BasicAgent initialized.")
     def __call__(self, question: str) -> str:
         print(f"Agent received question (first 50 chars): {question[:50]}...")
-        fixed_answer = "This is a default answer."
-        print(f"Agent returning fixed answer: {fixed_answer}")
-        return fixed_answer
+        answer = agent_run(question)
+        print(f"Agent returning answer (first 50 chars): {answer[:50]}...")
+        return answer
 
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """

main.py

@@ -1,6 +0,0 @@
-def main():
-    print("Hello from final-assignment-template!")
-
-
-if __name__ == "__main__":
-    main()

pyproject.toml

@@ -1,7 +0,0 @@
-[project]
-name = "final-assignment-template"
-version = "0.1.0"
-description = "Add your description here"
-readme = "README.md"
-requires-python = ">=3.14"
-dependencies = []

requirements.txt

@@ -1,2 +1,9 @@
-gradio
-requests
+gradio==6.11.0
+gradio[oauth]==6.11.0
+requests==2.33.1
+python-dotenv==1.2.2
+langchain
+langchain-core
+langchain-google-genai
+langgraph
+tavily-python