AgentSidekick

Computer Use

Controlling desktops, apps and shells like a person would. As of 2026-11-21, 296 indexed products match this subcategory (140 with full profiles); 143 were published in the last 90 days, against 56 in the 90 days before, roughly in step with the catalogue as a whole. By GitHub stars, Browser Use leads, holding 20% of the stars of the 254 products here with their own repository. The most common form is MCP servers (80% of indexed products).

SubcategorySegment label: Thin DER

Segment statistics

Whole catalogue (all 296 indexed products, not only profiled ones). Calculated 2026-11-21.

296DERIndexed products140 with full profiles
143DERLaunched, last 90 days56 in the 90 days before
39DERMomentum vs catalogue50 = same pace as the catalogue
20%DERLargest project's share of starsAcross 254 products with their own repository
0.085DERConcentration (HHI)About 12 equal-sized projects’ worth; unconcentrated
8SRCMedian starsPer product with its own repository
82%DEROpen sourceMarked open source; unknown counts as not

Thin. Among the smallest 30% of nodes at its level, and launches are not outpacing the catalogue. Size is ranked against the other subcategory nodes; momentum compares this node’s launches in the last 90 days with the 90 before, relative to the whole catalogue’s ratio (score 62 or more = faster, 38 or less = slower; below 5 dated launches it is not measured and the label uses size alone).

Concentration. The Herfindahl–Hirschman index (HHI) adds up each project’s squared share of stars. At 0.085, attention is spread as if across about 12 equally popular projects. Using the common competition-policy thresholds (0.15 and 0.25), that is unconcentrated. Stars measure attention, not usage or quality.

Which product forms serve it

The same need is met by MCP servers, agent skills, plugins, extensions and frameworks. This counts each form across the whole catalogue.

Products in Computer Use by product type, as of 2026-11-21
Product type IndexedShareWith a profile
MCP servers23780.1%86
Agent skills175.7%17
Agent projects134.4%13
Plugins103.4%6
Frameworks93.0%8
Infrastructure72.4%7
Libraries10.34%1
Automation platforms10.34%1
AI APIs10.34%1

4 capabilities

Indexed = whole catalogue; profiled = products with their own page. A product can sit in several, so the rows do not add up to the total.

CapabilityIndexed ProfiledNew, 90 daysLabel
Desktop ControlMouse, keyboard and app control on macOS, Windows or Linux.1174954Lagging the catalogue
Terminal & Shell AutomationRunning shell commands, SSH and terminal sessions.883343Lagging the catalogue
Robotic Process AutomationClassic RPA bots for repetitive UI work.19129Thin
Screen UnderstandingParsing screenshots into actionable UI elements.13129Emerging

Top 40 products by GitHub stars

Search the directory

Profiled products only (140 of 296), ranked by the stars of their own repository. Products that live in a subfolder of a larger repository (common for skills) show that whole repository's stars and are ranked after them; products without a repository come last. Stars are popularity, not quality or installs.

Computer Use: top profiled products by GitHub stars, 2026-11-21
#ProductTypePublisherCapabilities GitHub stars Last push LicenceSave and compare
1Browser UsePython library that lets LLM agents control a Playwright/CDP browser by reading page state and issuing actions, with a hosted cloud option.Librarybrowser-use122ktodayMIT
2anything-llmStop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience Agent projectMintplex-Labs68ktodayMIT
3cherry-studioYour AI workspace for models, agents, and productivity.Agent skillCherryHQ54ktodayAGPL-3.0
4herdrthe runtime your coding agents live onAgent projectherdrdev52k2d agoApache-2.0
5CuaOpen-source infrastructure for computer-use agents: sandboxed macOS, Linux and Windows VMs plus an agent SDK.InfrastructureCua33ktodayMIT
6OpenAI APIAPI for GPT and reasoning models via the Responses and Chat Completions endpoints, with built-in tools such as web search, file search, code interpreter and computer use.APIopenai32k1d agoApache-2.0
7OpenMausBotOpen-source Grok Bot alternative with a virtual machine that bots can useAgent projectmilind-soni26ktodayApache-2.0
8agent-desktopComputer use CLI for AI agentsAgent projectvercel-labs26ktodayApache-2.0
9SkyvernAutomates browser workflows using vision LLMs that read pages and act without site-specific selectors; self-hosted or cloud.Automation platformSkyvern23ktodayAGPL-3.0
10cc-haha本地优先的跨平台 Claude Code / AI Agent 桌面工作台:多 Agent 协作、Git Worktree、代码 Diff、技能市场、多模型、Computer Use、桌面宠物;支持微信、飞书、钉钉及 H5 等远程访问。 | Local-first Claude Code / AI Agent desktop workspace: multi-agent teams, Git worktrees, diffs, skills marketplace, multi-model, Computer Use, task-aware desktop pets; remote via WeChat, Feishu, DingTalk, Telegram, WhatsApp & H5.Agent projectNanmiCoder16k1d ago—
11Agent-SAgent S: the state-of-the-art open source computer use agentsAgent projectsimular.ai13ktodayApache-2.0
12Windows-MCPAn MCP Server for computer-use in Windows OSMCP serverCursorTouch12k1d agoMIT
13openrigBuild your own network of agents from Claude Code, Codex and Pi: persistent teams with roles, shared context and owned work.Agent skillmvschwarz8.9ktodayApache-2.0
14SteelOpen-source browser API for agents with session management, proxies and CAPTCHA handling, self-hosted or Steel Cloud.InfrastructureSteel8.0ktodayApache-2.0
15UpsonicBuild autonomous AI agents in Python.MCP serverUpsonic8.0k5mo agoMIT
16tuiosA terminal window manager that knows what your agents are doing. Tiling panes, workspaces, sessions that survive restarts, and one Inbox for every coding agent. Official only at tuios.dev and this repo. tuios has no X/Twitter account.Agent projectGaurav-Gosain6.6ktodayMIT
17MachinesStart, pause, resume, and run commands and file operations in Linux sandboxes.MCP serverpollinations5.3ktodayMIT
18Handsy AILet AI agents see and operate a real iPhone via an ESP32-S3 USB HID + AirPlay. No jailbreak.MCP serveronehumanbeing5.2ktodayMIT
19InteguruThe first AI agent that builds permissionless integrations through reverse engineering platforms' internal APIs.Agent projectInteguru-AI4.8k5mo agoAGPL-3.0
20arcboxRun AI agents on real and isolated machines — own kernel, filesystem, and network — with <100ms boot. Local first, OCI compatible, pure Rust.Agent projectarcboxlabs4.3k5d agoApache-2.0
21claude_codex_bridgeVisible multi-agent CLI workspace for mixing Codex, Claude, Gemini, Kimi, Qwen, Cursor, Copilot, Pi, OpenCode, and other AI coding agentsFrameworkSeemSeam3.6k1d ago—
22nanoMusenanoMuse: an open-source personal agent for every device you own. One agent with a name and a look of its own that does things, keeps working while the app is closed, remembers you, and asks before anything you cannot undo. On Android, iPhone and iPad, Windows / macOS / Linux and in the browser; your phone's screen and your computer as its hands.Agent skillnano-muse3.6ktodayGPL-3.0
23FuxiFuXi is a fast, self-contained AI coding agent that lives in your terminal — edit code, run commands, and drive tools, with cost-aware routing across LLM providers.Agent projectfuxicodex3.2k1mo ago—
24lcuCodex computer use, decoupled from the app, for usage inside any harness.Agent skillamontlabs2.4k2d agoMIT
25open-coworkOpen-source AI agent desktop app for Windows & macOS. One-click install Claude Code, MCP tools, and Skills — with sandbox isolation, multi-model support, and Feishu/Slack integration.Agent projectOpenCoworkAI2.3k1mo agoMIT
26solidworks-automation-skillReliable AI Skill + MCP toolkit for agent-driven desktop CAD automation.MCP serverwzyn200512162.2ktodayMIT
27bamaBAMA: Build Any Mac App. Agent skills to research, clone, design, build, optimize, and test native macOS apps with Swift and SwiftUI.Agent skilldaveebbelaar1.9k16d agoMIT
28OpenGUIOpenGUI is an Android GUI agent framework for phone-use AI that can see, plan, and operate real mobile apps through the GUI.FrameworkCore-Mate1.8k5d ago—
29pokiPoki uses your iPhone like you do. An open-source agent app that runs on your iPhone, reads the screen, and taps, types and swipes in real apps with your own model account.MCP serverJosssphZhou1.7ktodayAGPL-3.0
30ouroborosOuroboros — self-creating AI agent. Born Feb 16, 2026.Frameworkrazzant1.7ktodayMIT
31cli-agent-orchestratorMulti-agent orchestration for AI coding CLIs — Claude Code, Kiro, Codex, and more, coordinated in isolated tmux sessionsMCP serverawslabs1.6ktodayApache-2.0
32HarnessStart Claude Code and Codex agents the person can watch, wait for them and read their answers.MCP serverharness.autonomous.ai1.5ktodayMIT
33tty7The terminal that outlives its window. Shells survive quitting the app, agents can drive other agents, and remote machines work the same way. Pure Rust, GPU-rendered.Agent projectl0ng-ai1.5ktodayApache-2.0
34Agentlas-OSThe Open-Source AI Agent OS inspired by OpenAI Dots, Meta Muse, OpenClaw, Hermes & Mem0. Autonomous 3-tier headhunting, Receipt & Verifier loops, scoped memory, and BYOM for Claude Code, Codex & Cursor.Frameworkagentlas-ai1.4ktodayApache-2.0
35ClawBenchOpen-source benchmark for browser AI agents on daily tasks.InfrastructureTIGER-AI-Lab1.3k16d agoApache-2.0
36agent-vision-toolkit为纯文本模型"看图“设计更好的视觉工具箱和技能,支持多图理解,图片问答,前端UI还原、GUI 自动化等,并可选无缝接入多个主流agent,直接识别粘贴图片| A vision toolkit and skill designed for text-only llms — image Q&A, long-screenshot OCR, frontend UI restoration, and GUI automation, with optional seamless integration for Codex, Claude Code, Pi, Oh My Pi, and OpenCodeAgent skillAnionex1.3k1mo agoMIT
37cuga-agentCUGA is an open-source generalist agent harness for the enterprise, supporting complex task execution on web and APIs, OpenAPI/MCP integrations, composable architecture, reasoning modes, and policy-aware features.Agent projectcuga-project1.2k2d agoApache-2.0
38Cloudflare Sandbox SDKSDK for running isolated containers from Workers to execute code, run commands and manage files for agents.Infrastructurecloudflare1.2k11d agoApache-2.0
39claude-councilClaude Code plugin that asks several AI coding agents the same question and shows their answers side by side. Gemini, OpenAI, Grok, Perplexity, Kimi and any model OpenRouter routes to over API; codex, antigravity, grok, kimi and cursor CLIs on subscription auth; ollama locally. Codex specialists take coding tasks on their own branch (experimental).Pluginhex9514d agoMIT
40dsh-vision-toolkit[dsh]为纯文本模型设计更强大的视觉工具箱:一行安装使用、粘贴图片直接识别、多张图片问答、截图到前端UI 还原等|DeepSeek Harness-native integration for agent-vision-toolkit: image Q&A, long-screenshot OCR, UI restoration, grounding, pixel diff, Artifacts, and Web UI.Agent skillAnionex949todayMIT

Most common integrations

Third-party services named by products in this node, whole catalogue.

IntegrationProducts Share
Android124.1%
Playwright62.0%
Xcode / iOS51.7%
SAP20.7%
Telegram20.7%
X (Twitter)20.7%
Discord10.3%
GitHub10.3%
Slack10.3%
WhatsApp10.3%
Workday10.3%
n8n10.3%

Launches per month

New Computer Use products per monthSRC

Whole catalogue, by publication month, 2024-11 to 2026-11. The last month is partial; undated products are excluded.

New Computer Use products per monthWhole catalogue, by publication month, 2024-11 to 2026-11. The last month is partial; undated products are excluded.. Products published: 19 points, latest 39 products in 2026-11.
Show the data as a table
New Computer Use products per month. Whole catalogue, by publication month, 2024-11 to 2026-11. The last month is partial; undated products are excluded.
Products published (products)
2024-111
2025-033
2025-041
2025-071
2025-081
2025-097
2025-103
2025-124
2026-014
2026-028
2026-0315
2026-0419
2026-058
2026-0616
2026-0719
2026-0826
2026-0942
2026-1056
2026-1139

Related opportunities

Opportunity radar

Scored gaps for this node or anything beneath it. Scores rank hypotheses; they do not predict demand.

OpportunityCompetitionConfidenceScore
Coverage gap: Screen UnderstandingScreen UnderstandingMediumLow57.7
Coverage gap: Robotic Process AutomationRobotic Process AutomationHighLow56.7

Limits of these numbers

  • Products are placed here by whole-word keyword rules on their name, description and repository topics, and inherit every ancestor of a matched node. Products that do this without naming it are missed; ambiguous words cause false matches. Methodology
  • About half of the catalogue matches no node at all, so absolute counts understate every category.
  • Launch counts use the publication date each source reports; for auto-discovered repositories that is the repository’s creation date.
  • Stars and concentration cover only products with their own repository (not a subfolder of a larger one). They measure attention, not adoption.

Page data last updated .