agentsclimarketplace

Agnes generate video

Skill oh-summy/skills/skills/agnes-generate-video

Cross-agent AI skills for Claude, Codex, OpenCode, Kimi Code, and more. Install all at once or pick only what you need.

Install
npx -y skills add oh-summy/skills --skill agnes-generate-video

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Generate a video from a text prompt or an image using Agnes AI. Supports text-to-video and image-to-video via agnes-video-v2.0.

SKILL.md

3.5 KB, 839 tokens by cl100k_base, as published. Nobody here has run it

Agnes Generate Video

Generate videos with agnes-video-v2.0. Video generation is asynchronous; the script creates a task and polls until completion.

When to use

  • The user asks for a short video, animation, or cinematic clip.
  • The user provides a starting image and wants it animated.

Inputs

  • --prompt: Text description of the desired video.
  • --image: Optional starting image (local path or URL) for image-to-video.
  • --width, --height: Output resolution. Default 1152x768.
  • --duration: Target duration in seconds. Default ~5 seconds.
  • --frame-rate: FPS. Default 24.
  • --num-frames: Override duration; must follow 8n + 1 rule and be ≤ 441.
  • --poll-interval: Seconds between status checks. Default 10.
  • --max-wait: Max seconds to wait. Default 1800.
  • --download: Optional local path or directory to save the MP4. Default is ~/AgnesAI/videos/agnes_video_<timestamp>.mp4.
  • --no-open: Do not open the output folder after saving.

Prompt Enhancement (before running the script)

Consult the prompt-components skill to assemble a high-quality prompt:

  1. Determine interaction mode:
    • User gave subject + style + mood → Mode 1 (assemble directly, no questions)
    • User gave subject but missing style/mood → Mode 2 (suggest 2–3 combinations)
    • User is vague ("make a video") → Mode 3 (ask up to 4 questions)
  2. Read prompt-components/references/positive.md for styles and quality vocabulary.
  3. Read prompt-components/references/negative.md and auto-append relevant negatives (universal + video-specific + subject-category + style-conflict).
  4. Read prompt-components/references/motion.md for camera movement, pacing, and environment motion. Always specify at least one camera movement.
  5. Translate non-English user input to English for the final prompt.
  6. Keep positive prompt ≤ 80 words, negative ≤ 30 words.

Steps

  1. Enhance the prompt using the rules above (do not skip this).
  2. Run the helper script in this skill directory: 
    bash scripts/generate_video.sh --prompt "A cat walking on a beach at sunset" --duration 5
  3. Return to the user: both the saved file path and the video URL. The URL must always be shown even when the file is downloaded locally.

Example

bash scripts/generate_video.sh \
  --prompt "A cinematic drone shot over a futuristic city at dusk, neon lights, 4k quality" \
  --duration 10 \
  --download ./future_city.mp4

Output location

By default the completed MP4 is downloaded to ~/AgnesAI/videos/agnes_video_<timestamp>.mp4 and the folder is opened automatically.

  • macOS/Linux: ~/AgnesAI/videos/
  • Windows (Git Bash/WSL): ~/AgnesAI/videos/ resolves to your user profile folder.

Override with --download:

bash scripts/generate_video.sh --prompt "..." --download ./my_video.mp4
bash scripts/generate_video.sh --prompt "..." --download ./my_video_dir/

Skip opening the folder: add --no-open.

Templates / references

  • references/templates.md — prompt templates for cinematic text-to-video, image-to-video animation, product demos, social media shorts, keyframe transitions, etc.

API reference

Setup

Set AGNES_API_KEY as an environment variable or in a .env file.

What ships with it: 3 files

10.2 KB alongside SKILL.md, 1 of them executable

references/

scripts/

Gives 0 of the 12 instructions most video audio skills give in 839 tokens

Counted across 622 of the 795 authors here whose files we hold, read 2026-08-07

  • Read individual rule files for detailed explanationsin 21 of 622, across 10 files
  • Render final videoin 13 of 622, across 6 files
  • Use WAV PCM 16kHz mono audio formatin 12 of 622, across 3 files
  • Use this skill when dealing with Remotion codein 11 of 622, across 4 files
  • Save generated audio to a WAV filein 11 of 622, across 4 files
  • Handle conversion errors gracefullyin 10 of 622, across 6 files
  • Add captions to videos alwaysin 10 of 622, across 4 files
  • Generate music from text descriptions using MusicGenin 9 of 622, across 2 files
  • Do not skip pipeline layersin 9 of 622, across 3 files
  • Do not make one tool do everythingin 9 of 622, across 3 files
  • Use Azure Document Intelligence for complex PDFsin 9 of 622, across 4 files
  • Never ask the user to paste their full API keyin 9 of 622, across 3 files

Said here and by no other author read

  • consult the prompt-components skill before running
  • determine interaction mode from user input
  • read positive negative and motion reference files
  • auto-append relevant negative prompt terms
  • specify at least one camera movement
  • translate non-english input to english

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.