gitmyhub

awesome-ai-agents

★ 28k updated 1y ago

A list of AI autonomous agents

A curated directory of AI agents, software that autonomously completes multi-step tasks using AI. Covers open-source projects and commercial products with descriptions, categories, and links.

setup: easycomplexity 1/5

Awesome AI Agents is a curated directory of AI agents, where an agent is software built on a large language model that is meant to act on a goal (browse, write code, file tickets, place orders) rather than just answer a single question. The README welcomes the reader to the list and splits it into two parts: open-source projects, and closed-source projects and companies. Each entry has a name, a one-line description, a category (such as General purpose, Build your own, or Multi-agent), a longer feature list, and links to documentation, GitHub, Discord, or papers where relevant.

The repository is not code you install; it is a hand-curated reading list with a companion web version at e2b.dev/ai-agents that lets you filter by category and use case. Entries shown in the README portion include open-source projects like Adala (an autonomous data-labelling agent framework), Agent4Rec (a recommender-system simulator running a thousand LLM-powered agents seeded from the MovieLens dataset), AgentForge (a low-code platform for building and testing custom agents against several LLM providers), and AgentGPT (a browser-based no-code version of AutoGPT). Closed-source entries and many more open-source ones follow in the same format.

The list is kept alphabetical inside each section, and the project explicitly asks contributors to preserve that ordering and to use the submission form or a pull request to add new entries. The README also points readers at a separate companion list called Awesome SDKs for AI Agents, which catalogues frameworks and tools for building agents, as opposed to this list which is only for the agents themselves.

You would reach for this when you want a single place to browse what AI-agent products and open-source projects exist, for example when comparing tools that automate workflows, run multi-agent setups, or assist with a specific task. The README is maintained by the team behind e2b, who use it to also point readers at their own Code Interpreter SDK and cookbook. The repo's topic tags reinforce the scope: agent, ai, autonomous-agents, autogpt, babyagi, and copilot.

Where it fits