Zenbukko

Zenbukko is a local course archive and study-material processing toolkit. It downloads accessible NNN course lessons and materials, can transcribe media with whisper.cpp, and can extract text from PDFs into Markdown.

Documentation lives in docs/. Start there for architecture, usage, feature behavior, API contracts, and development rules.

Quick Commands

npm install
npm run type-check
npm run lint
npm test
docker compose config
docker compose build zenbukko-api zenbukko-web

Main Entrypoints

• CLI: zenbukko auth, zenbukko list-courses, zenbukko download, zenbukko download-all, zenbukko ocr-materials, zenbukko build-report-prompt, zenbukko setup-whisper, zenbukko transcribe.
• Servers: zenbukko api for Core API and zenbukko web for static UI/proxy.
• Web UI: run zenbukko api and zenbukko web, or docker compose --profile cpu up zenbukko-web.
• Docker data: bind ./data to /data; session defaults to /data/session.json; downloads default to /data/downloads.

Documentation Index

• docs/architecture/: module layout and data flow.
• docs/llm/: LLM-friendly index for cross-references and schema contracts.
• docs/usage/: CLI, web UI, Docker Compose, and outputs.
• docs/features/: sessions, chapter selection, materials, OCR, transcription, and report prompts.
• docs/decisions/: accepted behavior decisions.
• docs/reference/: configuration, API, file layout, troubleshooting.
• docs/development/: coding rules, tests, commits, verification.