Skip to content
Back to search

dream-to-video-skill

AI agent skill that transforms dream descriptions into cinematic videos — auto-generates prompts, submits to Jimeng via browser automation, and downloads finished videos with post-processing effects.

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/mediastormdev--dream-to-video-skill/badge.svg)](/listings/mediastormdev--dream-to-video-skill)

HTML

<a href="/listings/mediastormdev--dream-to-video-skill"><img src="/api/listings/mediastormdev--dream-to-video-skill/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.

Via skills CLI

npx skills add mediastormDev/dream-to-video-skill -s dream-to-video

Manual install

Clone the repo and symlink the skill directory into your agent's skill folder:

git clone https://github.com/mediastormDev/dream-to-video-skill.git
mkdir -p ~/.claude/skills
ln -s "$(pwd)/dream-to-video-skill/skills/dream-to-video" ~/.claude/skills/dream-to-video

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 · 99 lines

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

  1. # Dream-to-Video Skill
  2. [中文文档](./README.zh-CN.md)
  3. AI agent skill that transforms dream text into cinematic videos. Describe your dream, and the agent automatically generates professional video prompts, submits them to the [Jimeng](https://jimeng.jianying.com) platform via browser automation, downloads the finished videos, and applies post-processing effects.
  4. ## How It Works
  5. ```
  6. Dream text → AI transforms to cinematic prompt → Queue → Browser automation → Jimeng platform → Download → Post-processing → Video files
  7. ```
  8. 1. **Prompt transformation** — The AI agent converts your dream description into a detailed cinematic video prompt following 10 strict rules (photorealistic style, fisheye lens, no text overlays, silent narrative, etc.)
  9. 2. **Queue submission** — The prompt is added to a local task queue (SQLite + JSONL)
  10. 3. **Browser automation** — A background worker drives Chromium via Playwright to interact with the Jimeng video generation platform: login, submit prompts, upload reference images when needed, monitor progress
  11. 4. **Download & post-processing** — Completed videos are downloaded and automatically processed with an "Elliptic Shatter" edge effect (center-clear, shattered-glass edges), outputting both the original and the effect version
  12. ## Install
  13. ### Via skills CLI
  14. ```bash
  15. npx skills add mediastormDev/dream-to-video-skill -s dream-to-video
  16. ```
  17. ### Manual install
  18. Clone the repo and symlink the skill directory into your agent's skill folder:
  19. ```bash
  20. git clone https://github.com/mediastormDev/dream-to-video-skill.git
  21. mkdir -p ~/.claude/skills
  22. ln -s "$(pwd)/dream-to-video-skill/skills/dream-to-video" ~/.claude/skills/dream-to-video
  23. ```
  24. ## Requirements
  25. - Python >= 3.10
  26. - Chromium (installed via Playwright)
  27. - An API key from any supported provider:
  28. | Provider | Model | Get a key |
  29. |----------|-------|-----------|
  30. | **Claude** (default) | claude-sonnet-4 | [console.anthropic.com](https://console.anthropic.com/settings/keys) |
  31. | **OpenAI** | gpt-4o | [platform.openai.com](https://platform.openai.com/api-keys) |
  32. | **OpenRouter** | any model | [openrouter.ai](https://openrouter.ai/keys) |
  33. | **Google Gemini** | gemini-2.5-flash | [aistudio.google.com](https://aistudio.google.com/apikey) |
  34. ## Setup
  35. After installing the skill, clone this repo to get the Python toolchain:
  36. ```bash
  37. git clone https://github.com/mediastormDev/dream-to-video-skill.git
  38. cd dream-to-video-skill/dream_to_video
  39. pip install -r requirements.txt
  40. playwright install chromium
  41. ```
  42. Then log in to the Jimeng platform (one-time QR code scan):
  43. ```bash
  44. python main.py login
  45. ```
  46. ## Usage
  47. Once set up, just describe a dream to your AI agent:
  48. > "I dreamed I was running barefoot on a coastal boardwalk surrounded by hundreds of sea lions..."
  49. The agent will:
  50. 1. Transform your text into a cinematic video prompt
  51. 2. Submit it to the queue (`python main.py add "<prompt>"`)
  52. 3. Start the background worker (`python main.py worker`)
  53. 4. Notify you when the video is ready in `dream_to_video/output/`
  54. ### CLI Commands
  55. | Command | Description |
  56. |---------|-------------|
  57. | `python main.py login` | Log in to Jimeng via QR code |
  58. | `python main.py verify` | Check login status |
  59. | `python main.py add "<prompt>"` | Add a prompt to the task queue |
  60. | `python main.py worker` | Start background worker |
  61. | `python main.py status` | View task progress |
  62. | `python main.py generate "<prompt>"` | Single synchronous generation |
  63. | `python main.py serve` | Start FastAPI web server |
  64. ### Output
  65. Each task produces two video files in `dream_to_video/output/`:
  66. - `task_XXX_YYYYMMDD_HHMMSS.mp4` — Original video
  67. - `task_XXX_YYYYMMDD_HHMMSS_elliptic-shatter.mp4` — With shattered-glass edge effect
  68. ## License
  69. [MIT](./LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

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