skills.rest
zai-org
zai-orgOFFICIAL
Explore all skills by this author →
Website𝕏@Zai_orgGitHub

glmv-caption

Generate captions and summaries for images, videos, and documents via the GLM-V API.

458|40|Updated Mar 30, 2026
Search Tags:#image#multimodal#video#document#captioning#zhipu#glm-v
One-click install
npx skills add https://github.com/zai-org/GLM-skills --skill glmv-caption
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: glmv-caption
Source: https://github.com/zai-org/GLM-skills/tree/main/skills/glmv-caption
Command: npx skills add https://github.com/zai-org/GLM-skills --skill glmv-caption

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, and includes scripts (resource) components.

What problem does it solve?

Quickly obtain clear, detailed descriptions and summaries of visual and document content so users do not need to manually inspect or articulate the contents of images, videos, or files.

Core Features & Use Cases

  • Multimodal captioning: Generate natural-language captions for images, videos, and documents using the ZhiPu GLM-V API.
  • Flexible inputs: Accepts image URLs, local images encoded as base64, and file/video URLs with validation for formats and sizes.
  • Production-friendly output: Returns raw model outputs and token usage, supports streaming, custom prompts, and saves results to JSON for auditing.
  • Use Case: Content creators and accessibility teams can automatically produce image alt text, reporters can summarize video content, and researchers can extract visual summaries from batches of files.

Quick Start

Provide an image URL or upload an image and ask the skill to "Generate a detailed caption describing the visual content and salient elements."

Frequently Asked Questions about glmv-caption

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

FAQPage Schema
How do I generate captions for images and videos using an API?▼

To generate captions for images and videos, this Skill sends image URLs, base64 local files, or video URLs to the GLM-V API and returns natural-language descriptions. It handles multimodal inputs and outputs raw model text along with usage metadata.

Can I use local image files for multimodal captioning?▼

Yes, you can use local image files for multimodal captioning by encoding them as base64 strings. The Skill accepts base64 local images alongside image URLs and video URLs, enforcing image format and size limits during processing.

What do I need to authenticate document understanding and visual comparison tasks?▼

You need a ZHIPU_API_KEY to authenticate document understanding and visual comparison tasks. This key is required to access the GLM-V API, which processes document URLs and generates summaries or comparisons based on visual content.

Does this tool support batch processing for multiple image inputs?▼

Yes, this tool supports batch processing for multiple image inputs. It applies to single or multiple image URLs, enforcing model token and image-count constraints while returning raw model output and usage metadata for each batch.

What are the limitations when summarizing video content from URLs?▼

Limitations when summarizing video content from URLs include enforced image format and size limits, as well as model token and image-count constraints. The Skill returns raw model output and usage metadata without bypassing these API restrictions.

Related Skills

Explore More Design & Creative Skills→
ViewWayViewWay

glmv-grounding

Extract and visualize GLM-V grounding results from images or videos.

Community
Advanced
ibreez3ibreez3
21

get-feed-detail-skill

Retrieve Xiaohongshu feed metadata and comments via MCP HTTP API.

Community
Intermediate
NVIDIA-NeMoNVIDIA-NeMo
58

plugin-function

Create in-process request handlers for NeMo Platform plugins with CLI and HTTP interfaces.

Official
Advanced
davpatel605-beepdavpatel605-beep

llava

Analyze images and generate conversational responses using LLaVA vision-language models.

Community
Advanced