env.example

# Appwrite — keep endpoint and project ID in sync across both name pairs
NEXT_PUBLIC_APPWRITE_ENDPOINT=https://fra.cloud.appwrite.io/v1
NEXT_PUBLIC_APPWRITE_PROJECT_ID=63ef1792594ff4f50de2
APPWRITE_ENDPOINT=https://fra.cloud.appwrite.io/v1
APPWRITE_PROJECT_ID=63ef1792594ff4f50de2
APPWRITE_API_KEY=your-server-api-key-here

# Tables DB (env names use *_COLLECTION_ID for backward compatibility)
APPWRITE_DATABASE_ID=688f787e002c78bd299f
APPWRITE_POSTS_COLLECTION_ID=688f78a20022f61836ff
APPWRITE_VOTES_COLLECTION_ID=68910fc60001839e0ada
APPWRITE_COMMENTS_COLLECTION_ID=68993e6c003af0fe657f
APPWRITE_DAILY_TOPICS_COLLECTION_ID=daily_topics
APPWRITE_TOPICS_COLLECTION_ID=topics

# Next.js — canonical site URL (metadata, OG, absolute links)
NEXT_PUBLIC_BASE_URL=https://refetch.io

# OpenAI (Scout + Enhancement)
OPENAI_API_KEY=your-openai-api-key-here
# Optional; Enhancement reads this from global / project env (default gpt-4o-mini)
OPENAI_MODEL=gpt-4o-mini

# Scout function — comma-separated site roots to crawl
SCOUT_TARGET_WEBSITES=https://techcrunch.com,https://theverge.com,https://arstechnica.com,https://wired.com,https://venturebeat.com,https://engadget.com,https://infoq.com,https://news.ycombinator.com,https://lobste.rs,https://reddit.com/r/programming,https://sciencedaily.com,https://nature.com/nature/technology,https://9to5mac.com,https://9to5google.com,https://theregister.com,https://zdnet.com,https://www.technologyreview.com/tag/the-algorithm

# Readme function (GitHub + link base for generated README)
README_ORIGIN=https://refetch.io
README_GITHUB_TOKEN=your-github-pat-here
README_GITHUB_OWNER=refetch-io
README_GITHUB_REPO=refetch
README_GITHUB_BRANCH=main

# Optional — app/api/plausible/realtime
# PLAUSIBLE_API_KEY=your-plausible-api-key-here
# PLAUSIBLE_SITE_ID=your-site-id-here