deep-research / README.md
Vasilii Tokarev
Switch to chat-based UI
f3da3e8
|
raw
history blame
2.05 kB
metadata
title: Deep Research
emoji: πŸ†
colorFrom: red
colorTo: green
sdk: gradio
sdk_version: 5.34.0
app_file: src/main.py
pinned: false
license: apache-2.0

CD to HF Space

Deep Research

Deep Research is an AI-powered research assistant that helps users investigate complex topics by generating clarifying questions, planning web searches, and synthesizing detailed markdown reports.

Features

  • Clarification: Asks targeted questions to refine the research query.
  • Planning: Suggests up to five web searches to address the query.
  • Web Search: Summarizes web results for each planned search.
  • Report Generation: Produces a comprehensive markdown report with a summary and follow-up questions.
  • Interactive UI: Chat-based Gradio interface for end-to-end research.

Usage

  1. Send your research topic in the chat box.
  2. Answer the clarifying questions directly in the chat.
  3. Receive the detailed report back in the same conversation.

Tech Stack

  • Python, Gradio
  • openai-agents (for agent orchestration)
  • Async/await for non-blocking operations

File Overview

  • main.py β€” Gradio UI and app entry point
  • research_manager.py β€” Orchestrates the research workflow
  • research_agent.py β€” Core agent logic and tool coordination
  • clarifier.py β€” Generates clarifying questions
  • planner.py β€” Plans web search queries
  • web_search.py β€” Summarizes web search results
  • report_generator.py β€” Creates the final markdown report

Setup (with uv)

  1. Install Python 3.12 (if not already installed).
  2. Install uv:
    pip install uv
    
  3. Install dependencies:
    uv pip install -r pyproject.toml
    
  4. Run the app:
    uv pip run python src/main.py
    

The app will launch a Gradio interface in your browser.