MoonshotAI
Featured
Kimi-K3
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It shifts AI video prompting from generic cinematic keywords to a systematic framework that defines the 'identity' of the footage—who filmed it, with what device, and what went wrong—directly addressing the common 'too perfect, fake' problem in generated videos.
Who it is for
- Creators using text-to-video models (Seedance, Sora, Kling, Runway, Veo)
- Video prompt engineers who want realistic street photography, home video, or archival footage styles
- Short-form video producers needing consistent characters, timely actions, and ambient sound cues
- Developers integrating skill-based prompting into agents like Hermes, Claude, Codex, or Cursor
Use cases
- Generating a 15-second vertical smartphone clip of a New York morning commute with real exposure fluctuations and walking shake
- Creating a 2000s DV-cam home video of a Korean alley with compression artifacts, color fade, and autofocus hunting
- Converting a one-line idea into a full structured prompt with character description, location, camera defects, and a 10-15 second timeline
Strengths
- Structured prompt decomposition into 7 parts (character, location, visual style, camera style, timeline, audio, target)
- Built-in reference libraries for device-specific aesthetics (DV, VHS, Super8, smartphone, CCTV) and atmosphere translation
- Anti-AI checklist to enforce consistency, non-perfect events, and remove cinematic defaults
- Multiple installation options (Hermes skill, generic skill file, standalone methodology)
Considerations
- Requires familiarity with AI video generation tools and prompt engineering concepts
- Output is a structured text prompt, not a generated video; users still need a video model to execute
- Effectiveness depends on how well the underlying video model handles detailed technical constraints like sensor noise or autofocus behavior
README quick start
安装
Description
Hermes skill for realistic AI video prompts for Seedance and text-to-video models.
Related repositories
Similar projects matched by category, topics, and programming language.
xuchonglang
Featured
investing-for-beginners
A structured investing guide for Chinese beginners covering US stocks, options, and cryptocurrency, with focus on foundational concepts and risk awareness.
Krishnagangwal
Featured
CS-Fundamentals
A curated collection of Computer Science fundamentals (PDFs, notes, cheatsheets, interview question banks) for placement preparation, covering seven core subjects plus general resources.