README.md

metadata

title: NewsIntel Agent
emoji: 💻
colorFrom: indigo
colorTo: purple
sdk: gradio
app_file: app.py
python_version: 3.11
pinned: false
sdk_version: 5.42.0

🗞️ NewsIntel Agent — Job Briefings & Hiring Signals A Gradio app that pulls recent news, generates grounded summaries, detects entities & sentiment, and produces interview-ready briefings (HTML/PDF). Includes one-click modes (Healthcare / Finance / R&D), caching, charts, and optional S3/SES export + weekly email.

✨ Features Google/press-release RSS + your custom get_news source LLM-generated grounded summaries (via your services.llm.summarize) Entity extraction, keywording, sentiment Trend & simple 7-day forecast (linear fit on daily sentiment) Briefing export: HTML (built-in) and PDF (optional, via reportlab) Optional S3 upload + SES weekly email 📁 Repository layout (minimum) ├─ app.py # (rename your gradio_app -> app.py) ├─ requirements.txt ├─ README.md ├─ services/ │ ├─ init.py │ ├─ search.py # must define get_news(query, days, k) -> list[dict] │ ├─ llm.py # must define summarize(prompt) -> str │ ├─ huggingface_nlp.py # must define analyze_sentiment, analyze_entities, extract_keywords │ ├─ aws.py # optional: s3_upload(path), ses_send_email(to, subj, html) │ └─ cache.py # get_cache(*keys), set_cache(obj, *keys)

Do NOT upload your local .env. Use Spaces → Settings → Variables & secrets instead.

🔧 Environment variables (set in Spaces → Settings → Variables & secrets) Only set what you actually use in your services/ modules.

If your services/llm.py calls OpenAI (example):

OPENAI_API_KEY If your services/llm.py calls Hugging Face Inference API (example):

HF_TOKEN If enabling S3 uploads in services/aws.py:

AWS_ACCESS_KEY_ID AWS_SECRET_ACCESS_KEY AWS_DEFAULT_REGION (e.g., us-east-1) S3_BUCKET (bucket name) If enabling SES emails in services/aws.py:

SES_SENDER_EMAIL (verified in SES) (same AWS creds/region as above) If these aren’t set, the app will still run; S3/SES buttons will simply show fallback messages.

▶️ Run locally python -m venv .venv && source .venv/bin/activate # on Windows: .venv\Scripts\activate pip install -r requirements.txt python app.py