skills.rest
tmtgroupbot
tmtgroupbotCOMMUNITY
Explore all skills by this author →
GitHub

video-understand

Analyze video content, motion, and temporal sequences using the z-ai-web-dev-sdk vision API.

Updated Mar 30, 2026
Search Tags:#video analysis#content moderation#vision api#z-ai-web-dev-sdk#temporal analysis#action detection#video summarization
One-click install
npx skills add https://github.com/tmtgroupbot/lptracker-deploy --skill video-understand-tmtgroupbot
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: video-understand
Source: https://github.com/tmtgroupbot/lptracker-deploy/tree/main/skills/video-understand
Command: npx skills add https://github.com/tmtgroupbot/lptracker-deploy --skill video-understand-tmtgroupbot

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires z-ai-web-dev-sdk, and includes scripts (resource) components.

What problem does it solve? Analyzing video content manually is time-consuming, and most AI tools only handle static images. This Skill enables AI-driven understanding of video files, including scene description, action detection, event timelines, and content classification, through the z-ai-web-dev-sdk. ## Core Features & Use Cases - Video Scene & Motion Analysis: Describe scenes, detect actions, track people and objects across frames, and extract chronological event timelines from MP4, AVI, MOV, WebM, MKV, and FLV videos. - CLI and SDK Workflows: Use the z-ai CLI for quick one-off video summaries, or the SDK for multi-turn video conversations, batch processing, and production API integrations with Express.js or Next.js. - Use Case: A content platform needs to catalog hundreds of uploaded videos. Use this Skill to batch-process each video URL, generating summaries, tags, and moderation ratings automatically. ## Quick Start Ask the AI to analyze a video URL and summarize the main events, actions, and key moments in chronological order.

Frequently Asked Questions about video-understand

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze a video with the z-ai-web-dev-sdk?▼

Call zai.chat.completions.createVision with a message containing a text prompt and a video_url content block pointing to the video URL. The response contains the analysis in response.choices[0].message.content.

How to summarize a video using the z-ai CLI?▼

Run z-ai vision with the -p flag for your prompt and -i for the video URL or local file path, for example: z-ai vision -p "Summarize this video" -i "./meeting.mp4". Add -o to save the result as JSON.

What video formats does z-ai vision support?▼

Supported formats include MP4, AVI, MOV, WebM, MKV, and FLV. MP4 is the most widely supported and recommended format, and videos can be provided as public URLs or local file paths.

Can I use z-ai-web-dev-sdk in client-side browser code?▼

No, the SDK must be used in backend code only, never in client-side code. Import it in server-side environments such as Express.js endpoints or Next.js API routes to keep credentials secure.

Why is video analysis slow or missing temporal details?▼

Videos take longer to process than images, and complex temporal reasoning needs thinking mode enabled. For long videos, analyze shorter clips or sampled segments and write prompts specifying time segments like beginning, middle, and end.

Related Skills

Explore More Software Engineering Skills→
Pixazo-AIPixazo-AI

video-convert-api

Convert video files to MP4 format via the Pixazo API gateway.

Official
Basic
armoin2018armoin2018
4

Instagram Integration Manager

Publishes content, manages engagement, and collects analytics via Instagram Graph API.

Community
Advanced
thiansitthiansit

watch-visual

Analyze YouTube videos by aligning visual frames with audio transcripts.

Community
Intermediate
deepgramdeepgram
23

video-frames

Extract video frames at specific timestamps using ffmpeg.

Official
Basic