How hamzaamir1470/personal-ai-journalist works
Personal AI Journalist turns user-selected topics and source preferences into downloadable audio summaries. The Streamlit interface posts requests to a FastAPI endpoint, which conditionally runs news scraping and Reddit analysis, then generates a broadcast script and ElevenLabs audio for the frontend. The news path extracts headlines before summarization; the Reddit path uses an MCP-backed agent. README documentation is absent, so the map follows the sampled source; unsampled subsystems are not inferred.
Main entry points: backend.py Audio Endpoint · reddit_scrapper.py Reddit Analysis · frontend.py Topic Interface · news_scrapper.py News Scraper
Components 12
User Interface
- Topic Interface
frontend.py
Request Orchestration
- News Request
models.py - Audio Endpoint
backend.py
Source Analysis
- News Scraper
news_scrapper.py - News Extraction
utils.py - Reddit Analysis
reddit_scrapper.py
Broadcast Production
- Broadcast Script
utils.py - Audio Synthesis
utils.py
Other components
- Topic User
- ScraperAPI MCP
- Mistral AI
- ElevenLabs
Connections 13
- Topic User → Topic Interface: selects topics
frontend.py - Topic Interface → Audio Endpoint: posts request
frontend.py - Audio Endpoint → News Request: validates as
backend.py - Audio Endpoint → News Scraper: scrapes news
backend.py - Audio Endpoint → Reddit Analysis: scrapes Reddit
backend.py - News Scraper → News Extraction: extracts and summarizes
news_scrapper.py - News Extraction → Mistral AI: summarizes headlines
utils.py - Reddit Analysis → ScraperAPI MCP: loads tools
reddit_scrapper.py - Reddit Analysis → Mistral AI: runs agent
reddit_scrapper.py - Audio Endpoint → Broadcast Script: generates script
backend.py - Audio Endpoint → Audio Synthesis: converts script
backend.py - Audio Synthesis → ElevenLabs: requests speech
utils.py - Audio Endpoint → Topic Interface: returns audio
backend.py