The agents, and what they are good for
Profiles built from each vendor's docs, with the playbooks and lessons for each agent.
Coding agents
Anthropic's agent that works in your terminal, IDE, desktop app and browser. It reads and edits files, runs commands and connects to tools through MCP. People also use it for jobs far from code, such as bookkeeping, research and inbox work, because it can run scripts and keep notes in files.
OpenAI's coding agent. It runs as an open-source CLI, an IDE extension, inside the ChatGPT desktop app, and as Codex Cloud, which works on GitHub repos in its own cloud workspace. Developers use it to explore code, build features, fix bugs and review pull requests.
Cursor is an AI code editor and coding agent from Anysphere, which Cursor says SpaceX acquired in August 2026. It runs as a desktop editor, a CLI and cloud agents that work in their own VMs. Developers use it to plan and build features, fix bugs and review pull requests.
Gemini CLI is Google's open-source agent for the terminal, released under the Apache 2.0 license. It reads and edits files, runs shell commands, searches the web and calls tools through MCP. Google moved most individual users to Antigravity CLI in June 2026, so it now serves API key and Gemini Code Assist business users.
GitHub's AI coding assistant. It works in many IDEs, on github.com, in the Copilot CLI, the Copilot desktop app and GitHub Mobile. Developers use agent mode to edit code in the editor, and they assign issues to the cloud agent, which works in the background and opens a pull request.
Devin is an autonomous software engineering agent from Cognition. Each session gets its own workspace with a shell, an editor and a browser. Teams hand it tickets from Slack, Microsoft Teams, Linear or Jira, and it returns pull requests.
Jules is Google's asynchronous coding agent. It clones a GitHub repository into a cloud virtual machine, writes a plan for you to approve, then makes the change and offers it as a branch or pull request. People use it for bug fixes, tests, docs and small features while they do other work.
Lovable is an app builder that turns plain-language prompts into full web apps with a frontend, backend, database and sign-in. Lovable hosts the app and adds a built-in backend called Lovable Cloud. People use it for internal tools, SaaS prototypes, dashboards and marketing sites, and can sync the code to GitHub.
Replit Agent builds apps from a plain-language description inside Replit's online workspace. It sets up the project, writes the code, tests it and publishes it. People use it for web apps, mobile apps, dashboards, slides and automations without writing code themselves.
Assistants
Claude is Anthropic's general assistant, available on the web, in desktop apps for macOS and Windows, and on iOS and Android. People use it to write, analyze files, build documents and slides, and research questions. Connectors let it read and act in apps such as Google Drive, Slack and Linear, and the Research mode runs many searches before it answers.
OpenAI's AI assistant for the web, desktop and phone. It has a Chat mode for questions and drafts, and a Work mode for longer tasks that end in a file, analysis or report. People use it to research, write, analyze files and connect to tools such as Google Drive and Slack.
Gemini is Google's AI assistant app on the web, on phones and on the Mac. People use it to chat, research, write and work across Gmail, Docs, Calendar and other Google apps. Its agent feature, Gemini Spark, runs multi-step tasks and recurring schedules for paid subscribers.
Perplexity is an answer engine that searches the web and replies with cited sources. Paid plans add deeper Research reports, a choice of models from several labs, and file and app creation. People use it for quick fact finding, longer research and questions over their own files and connected apps.
Lindy is an AI teammate from Lindy.ai that you work with in Slack, by email, by text message and in a Gmail Chrome extension. It handles email triage and drafts, meeting recording and prep, scheduling, reports and CRM updates. Teams share one Lindy in Slack channels, and each person also has a private one.
Manus is a general AI agent that takes a goal and returns finished work, such as slides, websites, reports and data analysis. Each task runs on its own cloud computer with a browser, a file system and the ability to install software. Meta bought the company in late 2025, but Chinese regulators ordered the deal undone, and Manus said in August 2026 that it would run as an independent company again.
Grok is an AI assistant for chat, coding, image and video generation, and real-time answers from the web and X. It runs on grok.com, in iOS and Android apps and inside X. The Grok FAQ names SpaceXSI as the provider of the consumer service, while the developer API still runs under xAI.
Browser agents
ChatGPT agent was the agent mode in OpenAI's ChatGPT that took over from Operator. OpenAI's current docs no longer name it, and several guides report that ChatGPT Work replaced it in mid-2026. Today the same browser and form jobs run through Work, its cloud browser, the browser extension and Computer Use.
Claude in Chrome is Anthropic's extension for Google Chrome. It reads the page you are signed in to, then clicks, types, navigates and fills forms for you from a side panel. People use it for web tools that have no API, for background tasks across several tabs, and with Claude Code to test websites.
Comet is a Chromium-based web browser from Perplexity with a built-in assistant. The Comet Assistant sits in a side panel, reads the page you are on and can click, type and fill forms for you. People use it to summarize tabs, handle email and calendar, compare products and run multi-step web tasks.
Personal agents
Cowork is Anthropic's agent mode for knowledge work. It takes the Claude Code approach of planning and running multi-step tasks and applies it to files, spreadsheets, reports and research, with no terminal. Anthropic is now merging it into the main Claude app, so chat and Cowork share one message box and Claude decides when a request is a task.
Muse is Meta's personal AI agent, launched in the US in September 2026 and powered by Meta's Muse Spark model. It runs on its own cloud computer with a browser, and it can send email, book travel, fill out forms and buy things after you approve. It keeps working when the app is closed and messages you when something changes or needs your input.
OpenClaw is an open-source, MIT-licensed assistant that runs on your own hardware. A single Gateway process links chat apps such as WhatsApp, Telegram, Slack, Discord and iMessage to the model you choose. The OpenClaw Foundation, a US non-profit, stewards it, and people use it for inbox, calendar and other personal chores.
Hermes Agent is an open-source, MIT-licensed agent from Nous Research. It runs in a terminal or desktop app, and a gateway lets you reach it from Telegram, Discord, Slack, WhatsApp, Signal and email. It keeps memory across sessions and writes its own skills from tasks it has done.
Automation platforms
n8n is a workflow automation tool from the company of the same name. You build workflows on a visual canvas, and an AI Agent node lets a chat model pick tools inside a workflow. Teams use it to wire models such as OpenAI, Anthropic and Google into jobs that touch their other apps, either on n8n Cloud or on their own servers.
Zapier Agents lets you build AI agents that act in the apps Zapier connects to. You write instructions, set a trigger, add app actions and knowledge sources, then publish. Zapier is now moving Agents into AI by Zapier, a single AI step inside the Zap editor.
Make, formerly Integromat, is a visual automation platform. Its Make AI Agent (New) app puts an AI agent inside a scenario, with instructions, knowledge files and tools. People use it for jobs that need judgment on varied input, such as ticket triage, while fixed logic stays in normal scenario steps.