Deleting backend/extract.py

Deleting utils/helpers.py

Deleting config/__init__.py

Deleting backend/analyze.py

Deleting backend/scrape.py

Deleting utils/__init__.py

Deleting config/settings.py

Deleting backend/models.py

Deleting backend/app.py

Deleting backend/__init__.py

Deleting start.sh

Deleting README.md

Deleting requirements.txt

Deleting Dockerfile

Deleting .gitattributes

Put README.md

Put start.sh

Put Dockerfile

Put requirements.txt

Put utils/helpers.py

Put utils/__init__.py

Put config/settings.py

Put config/__init__.py

Put backend/analyze.py

Put backend/extract.py

Put backend/scrape.py

Put backend/models.py

Put backend/app.py

Put backend/__init__.py

Remove requirements.txt

Remove app.py

Put requirements.txt

Put app.py

Put app.py

Hybrid GLiNER2 + LFM pipeline with Analysis section

Add llama-cpp-python + gliner2 hybrid deps

Fix bs4 import typo

Fix Progress truthiness + suppress XML warning

Fix schema format for GLiNER2 extract_json

Update space description for GLiNER2

Replace LFM with GLiNER2 for structured JSON extraction

Replace llama-cpp with gliner2

Fix content stripping + auto-multi-page first pass

Add multi-page fetch with suggested internal links

Add Website/Extension mode selector with dual templates

Add structured template extraction

Add always-visible info panels

Update to factual extraction prompt, show system prompt in output

Show full extracted content, no limit

Show real extracted content in result