zai-org

zai-org / GLM-skills

Public

Official skills for the GLM family of models.

glm multimodal ocr skills vision
243
16
100% credibility
Found Apr 05, 2026 at 243 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A unified collection of command-line tools powered by GLM AI models for tasks like OCR, image generation, captioning, document conversion, resume screening, and stock analysis.

How It Works

1
🔍 Discover AI Helpers

You find a collection of simple AI tools that make everyday tasks like reading handwriting or generating images super easy.

2
🔑 Connect AI Service

Sign up for a free AI account and link it so the tools can think and create for you.

3
🛠️ Pick and Add a Tool

Choose a helper like 'read text from photos' or 'make image captions' and add it to your computer with one easy step.

4
📁 Feed Your File

Give the tool your photo, document, or video by dragging it in or typing a web link.

5
✨ Watch Magic Happen

The AI instantly analyzes your file and spits out perfect results like extracted text or smart descriptions.

🎉 Task Done Perfectly

You get exactly what you need – clean text, captions, or reports – ready to use right away.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 243 to 243 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

GLM-skills

GLM-skills is the official Python collection of CLI tools for Zhipu AI's GLM family, packing GLM 4.6 skills, GLM 4.7 flash skills, GLM 4.7 skills, GLM 5 skills, and GLM agent skills into agent-ready commands. It handles multimodal tasks like OCR for text/tables/formulas/handwriting, image generation, video/image captioning, object grounding with bounding boxes, PDF-to-PPT/web conversion, resume screening, and stock analysis. Devs get drop-in skills for Claude Code, OpenClaw, or custom agents via simple npx clawhub install or git clone, all powered by ZHIPU_API_KEY.

Why is it gaining traction?

Unlike scattered model repos, this unifies official GLM skills claude-style for agents, with CLI outputs in JSON for easy piping into workflows. Hooks include specialized GLM-OCR variants outperforming generics on docs/tables, grounding viz for images/videos, and PRD-to-app builders—saving hours on vision pipelines. Official GitHub CLI/actions integration and notebooklm agent skill links make onboarding fast.

Who should use this?

AI agent builders targeting GLM models for vision-heavy apps, like devs scripting OCR pipelines from screenshots/PDFs or grounding objects in security footage. Teams in finance using GLM stock analyst for HK/US reports, or product managers converting PRDs to web apps/prototypes. Claude Code/OpenClaw users needing GLM-specific multimodal boosts.

Verdict

Grab it if you're deep in GLM ecosystems—solid docs and CLI make it instantly useful despite 243 stars and 1.0% credibility signaling early maturity. Monitor official GitHub releases page for stability as adoption grows.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.