Generate videos using Grok AI via Chrome browser automation. Supports T2V (text-to-video) and I2V (image-to-video) with reference image uploads. Uses grok.com/imagine. Use when user says "Grok video", "create video with Grok", or wants AI video generation through Grok.
Installation
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Generate videos using Grok AI via Chrome browser automation. Supports T2V (text-to-video) and I2V (image-to-video) with reference image uploads. Uses grok.com/imagine. Use when user says "Grok video", "create video with Grok", or wants AI video generation through Grok.
Grok Video Generator
Generate videos using Grok AI via real Chrome browser automation (CDP) on grok.com/imagine.
Script Directory
Agent Execution Instructions:
Determine this SKILL.md file's directory path as SKILL_DIR
Replace all ${SKILL_DIR} in this document with the actual path
Script Reference:
Script
Purpose
scripts/main.ts
CLI entry point for Grok image & video generation
scripts/grok-utils.ts
Chrome CDP utilities (based on baoyu-post-to-x patterns)
Prerequisites
Google Chrome or Chromium
bun runtime
Logged in to x.com (session saved in Chrome profile)
Usage
Video Generation (uses grok.com/imagine)
# Text-to-video (T2V)
npx -y bun ${SKILL_DIR}/scripts/main.ts --video "A cat playing piano" --output cat.mp4
# Video with specific aspect ratio
npx -y bun ${SKILL_DIR}/scripts/main.ts --video --aspect 9:16 "Ocean waves crashing" --output waves.mp4
# Image-to-video (I2V) — animate a photo
npx -y bun ${SKILL_DIR}/scripts/main.ts --video \
-r /path/to/selfie.png \
--aspect 9:16 \
"Animate this photo. The woman turns toward camera with a confident smile and says hello." \
--output animated.mp4 \
--timeout 600
I2V Best Practices
Aspect
Recommendation
Aspect ratio
Always specify --aspect 9:16 for portrait, --aspect 16:9 for landscape
Timeout
Use --timeout 600 for video (generation takes 1-3 minutes)
Prompt style
Describe the motion/action you want, not just the scene
Reference image
Higher resolution = better quality output
Output resolution
Grok outputs ~416x752 for 9:16 — this is normal
Speech
Grok I2V does NOT generate real speech audio — for talking head videos with actual voice, use Seedance (--agent mode) instead
Options
Option
Description
Default
<text>
Generation prompt (positional)
—
--prompt, -p
Prompt text (alternative to positional)
—
--reference, -r
Reference image path (repeatable for multiple)
—
--output, -o
Output file path
grok-image.png
--all
Save all generated images (numbered: name-1.png, name-2.png)
false
--imagine
Use grok.com/imagine for image gen (clicks 📷 Image mode)
false
--video
Generate video (clicks 🎬 Video mode on grok.com/imagine)
Fixed in v2 (2026-03-21) — multi-strategy submit fallback. If still stuck, check Chrome profile login.
"Make video" button not found
Normal — Grok UI changes frequently. Script auto-falls back to submit button.
Low resolution video
Expected — Grok outputs ~416x752 for 9:16. Upscale with ffmpeg if needed.
Chrome won't launch
Check GROK_CHROME_PATH or ensure Chrome is installed at default path.
Session expired
Open Chrome manually, log in to x.com, close. Session persists in profile.
Changelog
v4 (2026-03-21): Silent failure detection:
Pre-existing image snapshot — captures all image URLs before submit, only counts NEW ones. Prevents returning avatars/sidebar thumbnails when generation silently fails.
Error detection — detects rate limits, captcha, "try again" messages from Grok and throws instead of returning garbage.
Stale video dedup — snapshots existing video URLs before submit, ignores them when polling. Fixes bug where previous generation's video was returned instead of new one.
