Agent TARS
by ByteDance
ByteDance's multimodal AI agent for desktop and browser automation — controls GUIs, browsers, and shell via vision and MCP tools. Available as npm CLI and desktop app.
Skills
GUI Control
Control desktop application interfaces using vision-based element detection, clicking, and keyboard input.
Browser Agent
Navigate websites, fill forms, click elements, and extract data from web pages autonomously via the browser.
Shell Execution
Execute shell commands and scripts as part of multi-step agent workflows with streamed output support.
MCP Tool Integration
Connect to local or remote MCP servers and invoke their tools as steps in multi-tool agent pipelines.
Related Agents
AgentOps for Apify Builders Bundle
Apify actor bundle for agent builders: normalize run traces for QA, control costs, and guard tool calls with a firewall…
BrowserOS
Drives a local open-source Chromium using your existing logged-in sessions, feeding external agents compact page snapsh…
Keenable
Serves a hundred-billion-document web index built for agents rather than human search pages, returning dense structured…
OpenSandbox
Runs AI-agent workloads in isolated Docker or Kubernetes sandboxes, exposing sandbox lifecycle, command, filesystem, an…