Repository: FujiwaraChoki/MoneyPrinter

Stars: 13104

CLAUDE.md Copy Chapter

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview Copy Chapter

MoneyPrinter automates YouTube Shorts creation from text topics. It uses Ollama for script generation, TikTok TTS for voiceover, Pexels for stock footage, and moviepy/ImageMagick for video composition. Output: a 9:16 vertical video (output.mp4).

Commands Copy Chapter

Setup Copy Section

cp .env.example .env        # then fill in API keys
uv sync                     # install dependencies
ollama serve                # start Ollama (separate terminal)
ollama pull llama3.1:8b     # pull default model

Run (local) Copy Section

uv run python Backend/main.py                              # API on :8080
uv run python Backend/worker.py                            # queue worker
python3 -m http.server 3000 --directory Frontend           # frontend on :3000

Run (Docker) Copy Section

docker compose up --build   # frontend :8001, backend :8080, postgres :5432

Verify Copy Section

uv run python -m compileall Backend          # syntax check
curl http://localhost:8080/api/models         # API smoke test

Tests Copy Section

uv run pytest -q                                           # all tests
uv run pytest tests/test_file.py::test_name -q             # single test

Architecture Copy Chapter

Video Generation Pipeline (end-to-end flow) Copy Section

User input (Frontend) → POST /api/generate → generation_jobs (Postgres queue)
  → worker.py claims queued job
  → gpt.py: generate_script() via Ollama
  → gpt.py: get_search_terms() → JSON keywords
  → search.py: Pexels API → download stock clips to temp/
  → tiktokvoice.py: TTS per sentence → MP3 chunks (threaded)
  → video.py: generate_subtitles() → .srt (AssemblyAI or local timestamps)
  → video.py: combine_videos() → concatenate/crop to 9:16
  → video.py: generate_video() → burn subtitles via ImageMagick, merge audio
  → (optional) mix background music from Songs/ at 10% volume
  → (optional) youtube.py: OAuth2 upload
  → output.mp4

Frontend ↔ Backend Communication Copy Section

Key Backend Modules Copy Section

Frontend Copy Section

Runtime Directories Copy Section

Required Environment Variables Copy Chapter

- TIKTOK_SESSION_ID — TikTok cookie for TTS
- PEXELS_API_KEY — stock video API
- IMAGEMAGICK_BINARY — leave empty to auto-detect from PATH

Optional: OLLAMA_BASE_URL, OLLAMA_MODEL, ASSEMBLY_AI_API_KEY, DATABASE_URL

Conventions Copy Chapter

- Python: 4-space indent, snake_case, type hints on all new/modified signatures, pathlib.Path for filesystem ops
- JS: camelCase, centralized API calls via apiRequest()
- API responses: {"status": "success|error", ...} with proper HTTP codes
- Long-running work: database-backed queue and separate worker process
- Concurrency: multiple jobs can be queued; worker processes them safely via DB locking
- Update docs/ when setup, env vars, or runtime behavior changes

README.md Copy Chapter

MoneyPrinter 💸

Sponsored by Post Bridge

<a href="https://www.post-bridge.com/?ref=moneyprinter">
<img src="docs/repo/PostBridgeBanner.png" alt="Post Bridge integration banner" width="720" />
</a>

---

𝕏 Also, follow me on X: @DevBySami.

Automate the creation of YouTube Shorts by providing a video topic.

MoneyPrinter is Ollama-first: script generation and metadata are fully powered by local Ollama models.

MoneyPrinter now uses a DB-backed generation queue (API + worker + Postgres in Docker) for reliable, restart-safe processing.

<a href="https://trendshift.io/repositories/7545" target="_blank"><img src="https://trendshift.io/api/badge/repositories/7545" alt="FujiwaraChoki%2FMoneyPrinter | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>

Important Please make sure you look through existing/closed issues before opening your own. If it's just a question, please join our discord and ask there.

🎥 Watch the video on YouTube.

Documentation Copy Chapter

Docs are centralized in docs/:

- Interactive Setup Script
- Quickstart
- Configuration
- Architecture
- Docker
- Testing
- Troubleshooting

FAQ 🤔 Copy Chapter

Which AI provider does MoneyPrinter use? Copy Section

MoneyPrinter is fully Ollama-based. Start Ollama, pull a model, and select the model in the UI.

ollama serve
ollama pull llama3.1:8b

How do I get the TikTok session ID? Copy Section

You can obtain your TikTok session ID by logging into TikTok in your browser and copying the value of the sessionid cookie.

My ImageMagick binary is not being detected Copy Section

MoneyPrinter auto-detects ImageMagick from your PATH on Linux, macOS, and Windows. If auto-detection fails, set the executable path manually in .env, for example:

IMAGEMAGICK_BINARY="C:\\Program Files\\ImageMagick-7.1.0-Q16\\magick.exe"

Don't forget to use double backslashes (\\) in the path, instead of one.

I can't install playsound: Wheel failed to build playsound: Wheel failed to build')" title="Copy section prompt for LLMs"> Copy Section

If you're having trouble installing playsound, you can try installing it using the following command:

uv pip install -U wheel
uv pip install -U playsound

If you were not able to find your solution, check Troubleshooting, ask in Discord, or create an issue.

Donate 🎁 Copy Chapter

If you like and enjoy MoneyPrinter, and would like to donate, you can do that by clicking on the button on the right hand side of the repository. ❤️
You will have your name (and/or logo) added to this repository as a supporter as a sign of appreciation.

Contributing 🤝 Copy Chapter

Pull Requests will not be accepted for the time-being.

Star History 🌟 Copy Chapter


License 📝 Copy Chapter

See LICENSE file for more information.