No visual example yet
Explore the skillAgent Vision Toolkit
Anionex
给纯文本 LLM agent 装上眼睛:图片问答、OCR、截图分析、视觉定位等一套视觉工具箱 + skill,并可无缝接入 Codex、Claude Code、OpenCode、Pi | Give text-only LLM agents vision: image Q&A, OCR, screenshot understanding,…
OPENAGENTSKILL / DIRECTORY
Temukan skill untuk tugas berikutnya dengan Codex, Claude Code, Cursor, dan lainnya.
5 Skills
Hasil: 5
No visual example yet
Explore the skillAnionex
给纯文本 LLM agent 装上眼睛:图片问答、OCR、截图分析、视觉定位等一套视觉工具箱 + skill,并可无缝接入 Codex、Claude Code、OpenCode、Pi | Give text-only LLM agents vision: image Q&A, OCR, screenshot understanding,…
No visual example yet
Explore the skillanthropics
Create original visual art, posters, PNG assets, and PDF documents through a clear design philosophy.
No visual example yet
Explore the skillapify
Crawlee—A web scraping and browser automation library for Node.js to build reliable crawlers. In JavaScript and TypeScript. Extract data for AI, LLMs, RAG, or GPTs. Down…
No visual example yet
Explore the skillapify
Crawlee—A web scraping and browser automation library for Python to build reliable crawlers. Extract data for AI, LLMs, RAG, or GPTs. Download HTML, PDF, JPG, PNG, and o…
No visual example yet
Explore the skillmicrosoft
Convert PDFs, Office documents, and web files into clean markdown for agents.