Skip to content
Back to search

vibevideo-mcp

Agent MCP for ffmpeg

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/hyepartners-gmail--vibevideo-mcp/badge.svg)](/listings/hyepartners-gmail--vibevideo-mcp)

HTML

<a href="/listings/hyepartners-gmail--vibevideo-mcp"><img src="/api/listings/hyepartners-gmail--vibevideo-mcp/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

You’ll need **Node.js**, **npm** (or pnpm/yarn), and **Python 3.12+** (with `pip`). These instructions assume a Unix-like OS (Mac/Linux), but should be adaptable for Windows. Docker instructions below.

- Start the Python backend first (main.py) - Then start the Node backend (npm run dev in ffmpeg-frontend)

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

readme.md · README · 158 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # VibeVideo-MCP
  2. ## Overview
  3. This monorepo contains an MCP server for agentic video editing. It also powers a front-end human user editor. It’s built and maintained by [HYE Partners](https://www.hyepartners.com).
  4. ---
  5. ## Installation
  6. You’ll need **Node.js**, **npm** (or pnpm/yarn), and **Python 3.12+** (with `pip`).
  7. These instructions assume a Unix-like OS (Mac/Linux), but should be adaptable for Windows.
  8. Docker instructions below.
  9. - Start the Python backend first (main.py)
  10. - Then start the Node backend (npm run dev in ffmpeg-frontend)
  11. - Then run the Python agent (ollamarun.py) for natural language jobs
  12. ---
  13. ## Clone the Repo
  14. - From your root coding project directory
  15. - git clone https://github.com/hyepartners-gmail/vibevideo-mcp.git
  16. - cd vibevideo-mcp
  17. ### Install the ffmpeg-backend components
  18. - cd ffmpeg-backend
  19. - python3 -m venv .venv
  20. - source .venv/bin/activate
  21. - pip install -r requirements.txt
  22. ### If you don't have ffmpeg installed
  23. - pip install ffmpeg
  24. ### Install the ffmpeg-frontend components
  25. - cd ffmpeg-frontend
  26. - npm install
  27. ---
  28. ## Structure
  29. The project includes **three main servers**:
  30. ### 1. React Front-End (Vite)
  31. - **URL:** http://localhost:8080/
  32. - **Dev:** Runs on Vite
  33. - **For:** User front-end that gives you access to all of the video editing tools
  34. - **Run:** (This launches *both* the React front-end and Node back-end in dev mode)
  35. - npm run dev
  36. ---
  37. ### 2. Node.js Back-End (Express)
  38. - **URL:** http://localhost:8300/
  39. - **Dev:** MCP Server, handles API requests from Python agents
  40. ---
  41. ### 3. Python Back-End (FFmpeg Command Runner)
  42. - **URL:** http://127.0.0.1:8200
  43. - **For:** Flask App handles all ffmpeg-powered media processing jobs
  44. - **Run:**
  45. - python main.py
  46. ---
  47. ### 4. Python Agent (Ollama Runner)
  48. - A simple Ollama agent running locally, for making one request at a time, media edits in natural language
  49. - **Run:**
  50. - python ollamarun.py
  51. **Suggested Models (must have function calling/tools):**
  52. - command-r7b:latest
  53. - devstral:latest
  54. - qwen3:latest
  55. - phi4-mini:latest
  56. - mistral-nemo:latest
  57. - llama3.1:8b
  58. - llama3.3:latest
  59. - qwen2.5-coder:latest
  60. - firefunction-v2:latest
  61. - llama4:scout
  62. ---
  63. ## Running with Docker (Optional)
  64. If you want everything to "just work" via Docker:
  65. Build and run both back-end servers with Docker Compose (from repo root):
  66. docker-compose up --build
  67. The following ports will be available:
  68. Frontend (Vite): http://localhost:8080/
  69. Node API server: http://localhost:8300/
  70. Python FFmpeg backend: http://localhost:8200/
  71. You can still use npm run dev/python main.py for local dev if you prefer.
  72. Note: The Docker setup mounts local source code, so edits will live-reload in most setups.
  73. ---
  74. ## Roadmap Items
  75. **June 3rd 2025**
  76. 1. Better Timeline editing, Get Render working
  77. 2. Dashboard Metrics / Run counts
  78. 3. Connector to CrewAI
  79. 4. MCP for some of the non-ffmpeg filters
  80. ---
  81. ## Usage
  82. - Free for any use: personal, research, or commercial.
  83. - **If you use this or build on top of it, a shout-out to [HYE Partners](https://www.hyepartners.com) is appreciated!**
  84. - **If you fix bugs please submit PRs so everyone can benefit**
  85. ## License
  86. Open use. Attribution requested, not required.
  87. For issues, improvements, or collaboration, visit [www.hyepartners.com](https://www.hyepartners.com).
  88. ---
  89. ## Common Problems & Fixes
  90. ### 1. Can’t find requirements.txt or Dockerfile?
  91. - These files are present in the repo under ffmpeg-backend/ and ffmpeg-frontend/.
  92. - If you don’t see them, make sure you’ve done a fresh git pull or git clone from the latest branch.
  93. - If issues persist, check your local .gitignore or .git/info/exclude for accidental blocks.
  94. ### 2. Flask or Other Python Dependency Errors?
  95. - If pip install -r requirements.txt fails with missing modules (like Flask), open requirements.txt and manually add any missing packages.
  96. - Run pip install -r requirements.txt again.
  97. ### 3. Node.js Import Error with JSON Files?
  98. - If you get SyntaxError: Unexpected identifier 'assert', you need "type": "module" in your package.json, and Node.js >= 17.5+.
  99. - For maximum compatibility, use require() instead of import for JSON in server files, or convert the file to CommonJS if needed.
  100. ### 4. Order of Startup
  101. - Python Backend (main.py) — must be running before the Node.js server.
  102. - Node.js Frontend/Backend (npm run dev) — launches both client and API server.
  103. - Python Agent (ollamarun.py) — optional, for natural language commands.
  104. - Docker Compose — launches all together.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.