Built with LangChain and Streamlit, this tool enhances your brainstorming experience by offering curated, optimized chains inspired by real-world effective brainstorming techniques. Rather than using ChatGPT directly, this app allows you to engage with structured brainstorming methods, guiding you to explore ideas comprehensively and maximize the benefits of LLM-driven brainstorming.
This app provides various brainstorming techniques:
- Big Mind Mapping: Expands ideas across a wide scope, ideal for when you need to gather a maximum number of ideas.
- Reverse Brainstorming: Identifies ways to create a problem, revealing potential pitfalls and fostering innovative solutions.
- Role Storming: Encourages adopting different perspectives to gather diverse insights.
- SCAMPER: Applies the SCAMPER technique (Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, Reverse) to prompt unique ideas.
- Six Thinking Hats: Based on Edward de Bono’s method, examines ideas from six angles: Data, Emotions, Risks, Benefits, Creativity, and Process Management.
- Starbursting: Generates questions using the 5 W's and 1 H (Who, What, Where, When, Why, How), offering in-depth topic exploration.
You can find the deployed app and test it directly here: https://brainstormers-2x7cxamrj2ngjbadgzthi4.streamlit.app/
The app’s pipeline is streamlined: instead of using agents or Langraph, it runs specific chains based on the selected brainstorming method. Each method initiates a structured chain tailored to the user’s choice. The output is organized into a tree of ideas, with an initial layer of ideas followed by deeper expansions as per each method's principles. For example, the Starbursting method generates detailed questions for each initial idea, followed by answers.
brainstomrmers.mp4
- Clone the repository:
git clone https://github.com/Azzedde/brainstormers
- Navigate to the project directory:
cd brainstormers - Install dependencies:
pip install -r requirements.txt
- Run the app:
streamlit run ./app.py
Future improvements include:
- Streamable output for real-time interaction
- Deployment on HuggingFace
- Integration with local LLMs
- Optimized runtime for Big Mind Mapping
This app is cost-effective, with Big Mind Mapping sessions costing around $0.01 per session, making it accessible and powerful for brainstorming needs.
This project represents the initial step toward creating a comprehensive tool, game, or drag-and-drop software with significant potential for businesses and individuals. To bring this vision to life, I'm seeking collaborators or sponsors who are interested in contributing to its growth. As I currently lack the budget and advanced software engineering skills to fully develop it on my own, your support could be invaluable. If you're interested in collaborating or sponsoring this project, please feel free to reach out!