Manus
General-purpose autonomous AI agent for multi-step tasks
Home / AI tools / AI Agents & Agent Frameworks
Explore autonomous AI agents, agentic browsers and the frameworks developers use to build multi-agent systems.
Agents pursue a goal across several steps, using tools such as browsers and APIs rather than answering a single prompt. Products range from ready-made assistants to developer frameworks for building your own.
Agents can be misled by instructions hidden in the pages and files they read. Limit their access, and start with tasks where a mistake is cheap and easy to undo.
A sensible starting shortlist from this list is Manus, Devin, Genspark. Run the same real task through each one before you pay for anything. Read how we select tools.
General-purpose autonomous AI agent for multi-step tasks
Autonomous AI software engineer from Cognition
AI agent workspace for research, slides and more
No-code AI agents for email, meetings and ops
Build an AI workforce of agents
OpenAI's browser with ChatGPT agent built in
AI-powered agentic web browser
AI browser from The Browser Company
Build custom agents for Microsoft 365
Google's platform for workplace agents
Autonomous agents across Salesforce
Custom AI agents connected to company knowledge
Marketplace and network of AI agents
Enterprise no-code AI agent builder
Agentic process automation for enterprises
Open-source library to let AI control browsers
Browser automation agents with computer vision
Framework for building LLM apps and agents
Stateful multi-agent orchestration framework
Framework for agents over your documents
Orchestrate role-playing AI agent teams
Microsoft's multi-agent framework
Lightweight framework for multi-agent workflows
Open standard for connecting AI to tools and data
Microsoft's SDK for AI agents in C#, Python and Java
Open-source framework for RAG and agents
Type-safe Python agent framework
Minimal code agents library from Hugging Face
TypeScript framework for AI agents
Framework and runtime for multi-agent systems
Stateful agents with long-term memory
Tool integrations for AI agents
Open-source LLM app and agent builder
Drag-and-drop builder for LLM agents
Visual builder for AI agents and RAG
Build and deploy conversational AI agents
Open-source platform for autonomous agents
AI agents focused on reliable software
Observability for AI agents
Build AI agents in natural language
Platform for autonomous AI agents
Open-source multi-agent framework
AI automation for teams
Enterprise AI agents
Enterprise AI assistant for employees
Agentic AI for service management
AI superconnector that introduces you to relevant people
Headless browser infrastructure for AI agents
Up-to-date library docs for AI coding agents via MCP
ByteDance's no-code AI agent and chatbot builder
Universal AI employee for enterprise workflows
Agentic AI browser that completes web tasks for you
Canvas-based AI agent workspace for long multi-step tasks
MCP server directory and AI workspace
Google's open-source Agent Development Kit
Open-source generalist AI agent platform (Suna)
Enterprise agent framework and agent studio
Open-source personal AI assistant that runs on your machine (formerly Clawdbot/Moltbot)
Opera's agentic AI browser
AI agent that operates your computer like a human
Team of AI employees for small business marketing and ops
Registry and hosting for MCP servers
Open-source browser API for AI agents
AWS's open-source SDK for building AI agents
AI browser with agent companions for research and sales tasks
AI agent that automates business tasks from plain-English instructions
Build AI agents that automate work across your apps
Personal cloud computer with an AI agent that runs tasks for you
This page lists 68 AI Agents & Agent Frameworks, reviewed for the November 2026 update. Discontinued and renamed products are removed or updated every month.
Many tools in this category offer a free plan, free credits, a trial or an open-source version. Plans change often, so confirm the current pricing on each tool's official website before you sign up.
Tools must have a live official website and a clear use case, and we favor established products and strong newcomers over copycats. Nobody can pay for a listing or a higher position. Read the editorial policy.