Prompt LibraryPrompt Library / Social Hook Clip (First-Frame Stopping Power) / for CogVideoX
Social Hook Clip (First-Frame Stopping Power) Prompt for CogVideoX
This is the CogVideoX version of the social hook clip (first-frame stopping power) prompt: a vertical clip engineered around the first frame: already mid-action at frame one, payoff held at the end — built to stop a scrolling thumb. It is formatted the way CogVideoX takes prompts best. Free — copy it, replace the bracketed parts, and paste it in.
The prompt
Push copies this prompt with a reference line naming your image, and downloads the image file — attach it in the model to complete the pair. This page stages the pair for you; it cannot upload into the model itself (sites cannot reach into each other), so the final attach happens there. The image is downscaled to at most 1024px and stored only in this browser, paired with this prompt. If browser storage is full, it is kept for this visit only.
The deliverable — and how to check it
Deliverable: one 6-second vertical 9:16 clip whose very first frame is already mid-action (no fade-in, no empty lead), ending on the payoff image held for the final full second. Acceptance check: scrub to frame one — if the action has not already started, the clip fails; the hook cannot be repaired in the edit.
Series: TikTok Production — Step 1 of 2 — stop the scroll. See its place in the full workflow
Using this prompt with CogVideoX
CogVideoX (Zhipu AI / THUDM) is an open-weights video model whose text encoder was trained on dense captions: it wants long, descriptive, caption-style sentences — a full paragraph describing the scene, the subject's appearance, the action and the camera, not a tag list. It does text-to-video and image-to-video at 720p, with no audio. The 2B checkpoint is openly licensed (Apache-2.0); the 5B carries the Tsinghua community license. Runs locally in Diffusers or ComfyUI.
What this route costs — and how to pay less
Cost structure: open weights — CogVideoX runs on your own hardware, so the marginal cost per generation is zero. The same prompt is pre-configured in this library for every major model, so you can run it on whichever route fits your budget — comparing the structures before you generate is a free way to save money.
Replace the bracketed placeholders — [HOOK ACTION], [SETTING], [PAYOFF IMAGE] — with your own details before pasting.
Other versions and customization
See all model versions of this prompt, or open the prompt gateway to build a custom prompt from your own one-line goal, formatted for CogVideoX or any other model.
More CogVideoX prompts
Mpromp2 — the AI Prompt Generator. Every prompt in this library is free to use, including commercially.