← Browser / computer-use

CogAgent

An open-sourced end-to-end VLM-based GUI Agent

AgentPick Score 40/100#52 in Browser / computer-use★ 1.2k GitHub starsApache-2.0Updated 12 Oct 2026Python

Ranked by AgentPick Score, our data score from public GitHub activity. Description is the project's own.

Source: github.com/zai-org/CogAgent

Topics: agent, computer-use, glm, gui-agent, vlm

Starter prompt

Using CogAgent, open [url], complete [task], take a screenshot after each step and report anything that failed.

Your tool here? Reach developers comparing agent tools.Get featured →

More in Browser / computer-use

Cua →

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and…

Stagehand →

The SDK to extract data and interact with any site on the web. Get started with Claude Code, Codex, Eve, Mastr…

CloakBrowser →

Stealth Chromium that passes every bot detection test. Drop-in Playwright replacement with source-level finger…

Midscene →

See interfaces like a human. E2E testing in natural language.

Skyvern →

Automate browser based workflows with AI

OpenCLI →

Make Any Website into CLI & Use your logged-in browser by AI agent.