# Autonomous Agents

> General-purpose agents that plan and execute multi-step tasks with minimal supervision.

Autonomous agents take a goal and plan and carry out the multi-step work to reach it — researching, calling tools, writing files, and acting with little supervision. This is the most exciting and least settled category: the gap between a slick demo and reliable real-world work is still wide, so weigh how dependably an agent finishes long, open-ended tasks over how impressive its highlight reel looks. The decisions that matter are reliability on realistic work, the breadth and safety of the tools it can use, and how much visibility and control you get — can you watch its steps, interrupt it, set budgets, and stop runaway loops before they run up a bill? Cost is a genuine design constraint: agents that plan, act, and observe in a loop can burn a lot of tokens, so predictable spend and step limits are features, not afterthoughts. Prefer tools that are transparent about failure modes and that keep a human in the loop for consequential actions. Because standards here are thin and the field turns over quickly, a recent track record and active maintenance matter more than in almost any other category. Start with bounded, well-defined tasks and human review, then widen the leash only as an agent earns it on your own workload — reliability compounds, and so do mistakes.

21 tools indexed. Sponsored placements are labeled and never ranked; open-source ordering reflects real GitHub adoption. Browse on the web: https://agentsearchengine.app/category/autonomous-agents

## Commercial products (alphabetical — stars don't measure commercial adoption)
- [Genspark](https://agentsearchengine.app/agents/genspark) — Super-agent that researches, calls, and creates on your behalf. · Freemium · Agent · best for consumer super-agent tasks: research, calls, and content creation
- [Manus](https://agentsearchengine.app/agents/manus) — General AI agent that turns your thoughts into actions across real tasks. · Freemium · Agent · best for delegating general multi-step tasks to a hosted autonomous agent

## Open-source index (ranked by adoption)
1. [AutoGPT](https://agentsearchengine.app/agents/autogpt) — AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters. · Open source · Agent · 186k★ · Python · active today · best for experimenting with long-running autonomous agent loops in an open-source stack
2. [browser-use](https://agentsearchengine.app/agents/browser-use) — Make websites accessible for AI agents. Automate tasks online with ease. · Open source · Agent · 108k★ · Python · MIT · active today · best for developers giving agents real browser control via an open-source python library
3. [ragflow](https://agentsearchengine.app/agents/ragflow) — RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs · Open source · Platform · 87k★ · Go · Apache-2.0 · active today · best for production rag over complex documents with agent capabilities, self-hosted
4. [deer-flow](https://agentsearchengine.app/agents/deer-flow) — An open-source long-horizon SuperAgent harness that researches, codes, and creates. · Open source · Agent · 79k★ · Python · MIT · active today · best for orchestrating sub-agents, memory, and sandboxes through one extensible skill system
5. [anything-llm](https://agentsearchengine.app/agents/anything-llm) — Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience · Open source · Platform · 64k★ · JavaScript · MIT · active 3 days ago · best for a private, all-in-one ai workspace with rag and agents you can run locally
6. [goose](https://agentsearchengine.app/agents/goose) — an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM · Open source · Agent · 52k★ · Rust · Apache-2.0 · active today · best for an extensible open-source agent on your machine that goes beyond coding tasks
7. [nanobot](https://agentsearchengine.app/agents/nanobot) — Lightweight, open-source AI agent for your tools, chats, and workflows. · Open source · Agent · 47k★ · Python · MIT · active today · best for a small, readable self-hosted agent framework with memory and mcp built in
8. [CowAgent](https://agentsearchengine.app/agents/cowagent) — Open-source super AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-model, multi-channel. Lightweight, extensible, one-line install. · Open source · Agent · 46k★ · Python · MIT · active today · best for an always-on personal assistant that controls your computer and chat platforms
9. [agno](https://agentsearchengine.app/agents/agno) — Build, run, and manage agent platforms. · Open source · Agent · 42k★ · Python · Apache-2.0 · active today · best for running a self-owned agent platform with rbac, a runtime, and a management ui
10. [wshobson/agents](https://agentsearchengine.app/agents/wshobson-agents) — Multi-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLI · Open source · Platform · 38k★ · Python · MIT · active 12 days ago · best for a prebuilt library of agents and skills shared across claude code, cursor, and codex
11. [agenticSeek](https://agentsearchengine.app/agents/agenticseek) — Fully Local Manus AI. No APIs, No $200 monthly bills. Enjoy an autonomous agent that thinks, browses the web, and code for the sole cost of electricity. · Open source · Agent · 27k★ · Python · GPL-3.0 · active 2 days ago · best for a fully local manus alternative that keeps browsing and coding on your own hardware
12. [pentagi](https://agentsearchengine.app/agents/pentagi) — Fully autonomous AI Agents system capable of performing complex penetration testing tasks · Open source · Agent · 22k★ · Go · MIT · active 2 days ago · best for autonomous penetration-testing agents with supervision and multi-provider llm support
13. [eigent](https://agentsearchengine.app/agents/eigent) — Eigent: The Open Source Cowork Desktop to Unlock Your Exceptional Productivity. Local and Free Alternative to Claude Cowork. · Open source · Agent · 15k★ · TypeScript · Apache-2.0 · active 2 days ago · best for a desktop multi-agent workforce you can run fully standalone
14. [LobsterAI](https://agentsearchengine.app/agents/lobsterai) — Open-source, desktop-grade AI agent that gets real work done — data analysis, slides, docs, video & web research. · Open source · Agent · 5.8k★ · TypeScript · MIT · active today
15. [moltis](https://agentsearchengine.app/agents/moltis) — A secure persistent personal agent server in Rust. One binary, sandboxed execution, multi-provider LLMs, voice, memory, Telegram, WhatsApp, Discord, Teams, and MCP tools. · Open source · Agent · 2.8k★ · Rust · MIT · active today
16. [codel](https://agentsearchengine.app/agents/codel) — Fully autonomous AI Agent that can perform complicated tasks and projects using terminal, browser, and editor. · Open source · Agent · 2.5k★ · TypeScript · AGPL-3.0 · active 2 years ago
17. [agi](https://agentsearchengine.app/agents/agi) — The first distributed AGI system. Thousands of autonomous AI agents collaboratively train models, share experiments via P2P gossip, and push breakthroughs here. Fully peer-to-peer. · Open source · Agent · 2.0k★ · MIT · active today
18. [agentic_security](https://agentsearchengine.app/agents/agentic-security) — Agentic LLM Vulnerability Scanner / AI red teaming kit · Open source · Agent · 1.9k★ · Python · Apache-2.0 · active 3 days ago
19. [zeroshot](https://agentsearchengine.app/agents/zeroshot) — Your autonomous engineering team in a CLI. The agent loop produces senior-level code that you can actually trust in prod because of non-negotiable feedback from independent reviewers. · Open source · Agent · 1.7k★ · JavaScript · MIT · active today

## Frequently asked
**What is an autonomous AI agent?**
An autonomous agent is given a goal and then plans and executes the steps to reach it — calling tools, browsing, and making decisions — with minimal human input.

**How do I evaluate an autonomous agent?**
Look at task-completion reliability on realistic work, the breadth of tools it can use, transparency (can you see and interrupt its steps), and safeguards against costly mistakes.

**Are autonomous agents production-ready?**
For narrow, well-defined tasks, increasingly yes; for open-ended work they still need oversight. Start with bounded tasks and human review.

---
Source: [Agent Search Engine](https://agentsearchengine.app). Full directory in one file: https://agentsearchengine.app/llms-full.txt