README.md

metadata

license: apache-2.0
title: YouTube Competitor Analysis
sdk: gradio
emoji: 🚀
colorFrom: red
colorTo: yellow
pinned: false
short_description: This application analyzes competitor YouTube channels and de

YouTube Competitor Analysis (Global)

This application analyzes competitor YouTube channels and detects global trending clusters of people mentioned in videos using Gemini (Google Generative AI) and the YouTube Data API.

This README explains how to run the app and where to enter your API keys in the UI.

Key points

• API keys are entered via the Gradio UI (Data Update tab) — they are not read automatically from environment variables.
• Keys are stored in memory for the running session only (not persisted to disk).

Requirements

The project has a requirements.txt. From your project root on Windows PowerShell run:

python -m pip install -r .\requirements.txt

If you don't have a virtual environment, consider creating one first:

python -m venv venv
.\venv\Scripts\Activate.ps1
python -m pip install -r .\requirements.txt

Run the app

From the repository root run:

python .\app.py

Gradio will print a local URL (and possibly a public share URL). Open the local URL in your browser.

Where to enter API keys

1. Open the Gradio UI and go to the "🔄 Data Update" tab.
2. Enter your keys in the two password-style text boxes:
   • YouTube API Key — for the YouTube Data API (developerKey)
   • Gemini API Key — for the Google Generative AI client
3. Click Apply API Keys. The status box will show whether configuration succeeded.

After applying keys you can use:

• Start Global Data Update to fetch video data and run person-detection logic.
• Channel management features (Add/Delete/Update) which require a configured YouTube client.

Behavior & notes

• Keys are kept in memory for the running Gradio session. If you stop the app you must re-enter them.
• If you see messages like YouTube API key is not set. or AI features will not work, confirm you applied the keys successfully.
• The app does not currently persist keys to disk. If you want persistence (encrypted local store or OS keyring) I can add that.

Troubleshooting

• Missing dependencies: re-install with pip install -r requirements.txt.
• If the YouTube API returns quota errors, check your API quota in Google Cloud Console.
• If Gemini calls fail, ensure the Gemini key is valid and that the google-generativeai client package version in requirements.txt is compatible.

Next steps (optional)

• Persist keys securely between sessions (encrypted file or OS keyring).
• Pre-fill UI inputs from environment variables for dev convenience.
• Add unit tests to mock Gemini/YouTube clients and verify wiring.

If you'd like one of those, tell me which and I will implement it.