AI Brochure Maker

Create a polished company brochure draft from a website. The app scrapes the site, filters for relevant pages (e.g., About, Careers), and uses an LLM (via OpenRouter) to produce a well-structured Markdown brochure.

---

Impact

This project demonstrates the power of combining web scraping, intelligent link filtering, and Al-driven content generation to deliver ready-to-use marketing assets. It shows how businesses can reduce costs, speed up content creation, and maintain consistent messaging with minimal human input.

---

🛠️ Tech Stack

• Python
• Streamlit (app UI)
• Libraries:
  • requests – fetch web content
  • openai – connect to LLM via OpenRouter
  • beautifulsoup4 – HTML parsing

---

Features

• Scrapes landing page and finds relevant links (About, Careers, Team, Values, etc.)
• Caches network calls to keep the app fast
• Two-column layout: status/raw data vs. final brochure
• Friendly error handling and progress spinners
• Website Scraping: Fetches company web pages with proper request headers.
• Content Filtering: Identifies relevant links such as About, Company, or Careers.
• AI-Powered Summarization: Uses an LLM (via OpenAI/OpenRouter API) to condense content into brochure-ready text.
• Structured Output: Formats results as Markdown for brochure sections.
• Interactive App: Streamlit UI for step-by-step brochure creation and refinement.

---

Setup

1) Install Dependencies

pip install -r requirements.txt

2) Configure Secrets

Create .streamlit/secrets.toml and add your OpenRouter API key:

Openrouter_Api_Key="sk-or-..."

Alternatively, you can set the environment variable OPENROUTER_API_KEY.

3) Environment Variables

• OPENAI_API_KEY (optional; not required if using OpenRouter via secrets)

4) Run Locally

streamlit run streamlit_app.py

Open the printed local URL in your browser.

---

Deployment

• Push this repository to GitHub
• On Streamlit Community Cloud, create a new app, point it to your repo, and set up .streamlit/secrets.toml with Openrouter_Api_Key
• Deploy. The app will use secrets in the cloud environment

---

Notes

• Be sure the target site allows scraping and respect robots.txt and legal terms
• If few relevant links are found, try more specific subpages or different domains

---

Example Workflow

• Input: https://examplecompany.com
• Extracted: About Us, Services, Careers pages
• Output: A company brochure summarizing mission, values, offerings, and opportunities.

---

🔮 Future Improvements

• Add image/logo scraping for branding.
• Export brochures as PDF or Word documents.
• Support multi-language brochure generation.