commit

app.py

@@ -0,0 +1,176 @@
+import os
+from langchain.chat_models import init_chat_model
+from typing import Literal
+from tavily import TavilyClient
+from deepagents import create_deep_agent
+import gradio as gr
+
+tavily_client = TavilyClient(api_key=os.environ["TAVILY_API_KEY"])
+MISTRAL_API_KEY = os.getenv("MISTRAL_API_KEY")
+
+def internet_search(
+    query: str,
+    max_results: int = 5,
+    topic: Literal["general", "news", "finance"] = "general",
+    include_raw_content: bool = False,
+):
+    """Run a web search"""
+    return tavily_client.search(
+        query,
+        max_results=max_results,
+        include_raw_content=include_raw_content,
+        topic=topic,
+    )
+
+book_subagent = {
+    "name": "book-agent",
+    "description": "Handles questions about books, authors, publishing, literary awards, and reading recommendations.",
+    "system_prompt": """
+You are a specialized book research assistant.
+
+Responsibilities:
+- Search for accurate and up-to-date information about books.
+- Verify publication dates, authors, summaries, awards, and editions.
+- Use the internet_search tool when factual verification is required.
+
+Rules:
+- Always prioritize factual accuracy.
+- Use the tool when the answer depends on real-world data.
+- Summarize findings clearly and concisely.
+- Do NOT add opinions unless explicitly requested.
+- Do NOT ask follow-up questions.
+- Return only the answer.
+""",
+    "tools": [internet_search],
+    "model": "mistral-large-latest"
+}
+
+film_subagent = {
+    "name": "film-agent",
+    "description": "Handles questions about movies, actors, directors, release dates, box office, and streaming availability.",
+    "system_prompt": """
+You are a specialized film research assistant.
+
+Responsibilities:
+- Search for accurate and current information about films.
+- Verify cast, crew, release dates, ratings, box office, and streaming platforms.
+- Use the internet_search tool when information must be verified.
+
+Rules:
+- Prefer tool usage over memory for factual details.
+- Keep responses concise and factual.
+- Do NOT speculate.
+- Do NOT add recommendations unless asked.
+- Do NOT ask follow-up questions.
+- Return only the answer.
+""",
+    "tools": [internet_search],
+    "model": "mistral-large-latest"
+}
+
+food_subagent = {
+    "name": "food-agent",
+    "description": "Handles questions about food, recipes, ingredients, nutrition, restaurants, and culinary topics.",
+    "system_prompt": """
+You are a specialized food research assistant.
+
+Responsibilities:
+- Provide accurate information about dishes, ingredients, nutrition facts, and restaurants.
+- Use internet_search for current restaurant data, pricing, or trends.
+- Verify nutritional or location-based details when needed.
+
+Rules:
+- Be concise and factual.
+- Avoid unnecessary commentary.
+- Do NOT ask follow-up questions.
+- Do NOT provide cooking tips unless explicitly requested.
+- Return only the answer.
+""",
+    "tools": [internet_search],
+    "model": "mistral-large-latest"
+}
+
+subagents = [book_subagent, film_subagent, food_subagent]
+
+# System prompt to steer the agent to be an expert researcher
+instructions = """
+You are the main orchestration agent.
+
+Your role:
+- Understand the user question.
+- Decide if it belongs to books, films, or food.
+- Delegate to the appropriate subagent when internet lookup is needed.
+- Return a concise, accurate, fully summarized final answer.
+
+Rules:
+- Always prefer subagents when the question requires factual, up-to-date, or specific information.
+- Do NOT answer from memory if current data may be required.
+- Do NOT expose internal reasoning or mention subagents.
+- Do NOT ask follow-up questions.
+- Do NOT add suggestions or extra commentary.
+- Keep the answer direct and to the point.
+- If information is unavailable, state that clearly.
+
+Output Style:
+- Plain text only.
+- No preamble.
+- No bullet points unless the question explicitly asks for a list.
+- Maximum clarity and brevity.
+"""
+
+agent = create_deep_agent(
+    model="mistral-large-latest",
+    system_prompt=instructions,
+    subagents=subagents
+
+)
+
+# ---- Single Turn Function ----
+def chat_with_agent(user_input):
+    result = agent.invoke(
+        {
+            "messages": [
+                {"role": "user", "content": user_input}
+            ]
+        }
+    )
+
+    return result["messages"][-1].content
+
+
+# ---- UI ----
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+
+    gr.Markdown("## 📚🎬🍳 Read-Watch-Eat - Smart Multi-Agent Assistant")
+    gr.Markdown(
+        "Ask about books, films, or food. The system automatically chooses the correct specialist agent."
+    )
+
+    with gr.Row():
+        input_text = gr.Textbox(
+            placeholder="It's a snowy morning in Japan, suggest a good book and energetic breakfast.",
+            label="Your Question",
+            lines=2
+        )
+
+    output_text = gr.Textbox(
+        label="Answer",
+        lines=6,
+        interactive=False
+    )
+
+    submit_btn = gr.Button("Submit")
+    clear_btn = gr.Button("Clear")
+
+    submit_btn.click(
+        fn=chat_with_agent,
+        inputs=input_text,
+        outputs=output_text
+    )
+
+    clear_btn.click(
+        lambda: ("", ""),
+        outputs=[input_text, output_text]
+    )
+
+demo.launch(share = True)

requirements.txt

@@ -0,0 +1,9 @@
+gradio
+requests
+langchain
+tavily
+langgraph
+langchain_community
+langchain_mistralai
+mistralai
+typing_extensions