Picture by Editor
# Introduction
If you wish to be taught agent engineering by doing as an alternative of simply studying about it, the easiest way remains to be to fork actual repos, run them regionally, and alter them to your personal use. That is the place the true studying occurs. I’ve hand-picked the very best 10 ones, the tasks which can be each helpful and well known, so you possibly can see how agent apps are being constructed as we speak. So, let’s get began.
# 1. OpenClaw
OpenClaw (~343k ⭐) is the one I’d level to first if you wish to see what the subsequent wave of non-public AI assistants would possibly appear like. It’s constructed as a private assistant that runs by yourself gadgets and connects to the instruments folks already use, like WhatsApp, Telegram, Slack, Discord, Sign, and iMessage. What makes it attention-grabbing is that it isn’t only a easy chat demo. It looks like an actual assistant product, with multi-channel help, voice options, and a broader ecosystem round expertise and management. If you’d like a repo that feels near an actual agent system, it is a robust place to start out.
# 2. OpenHands
OpenHands (~70k ⭐) is a good repo to fork in case your essential curiosity is coding brokers. It’s constructed round AI-driven improvement and now has a wider ecosystem round it, together with cloud, docs, CLI, SDK, benchmarking, and integrations. That issues as a result of you aren’t simply taking a look at one demo. You may research the core agent, test the interface, and likewise see how the staff thinks about analysis and deployment. If you wish to construct or customise a coding assistant, this is without doubt one of the most sensible repos to be taught from.
# 3. browser-use
browser-use (~85k ⭐) is without doubt one of the most helpful tasks in order for you brokers that may truly do issues on the internet. The concept is easy: it makes web sites simpler for AI brokers to make use of, to allow them to deal with browser-based duties with much less friction. That makes it straightforward to experiment with, since lots of actual agent work leads to the browser anyway — type filling, analysis, navigation, and repetitive on-line duties. It additionally has supporting repos and examples, which makes it simpler to go from curiosity to one thing you possibly can check in an actual workflow.
# 4. DeerFlow
DeerFlow (~55k ⭐) is without doubt one of the extra attention-grabbing tasks if you wish to perceive long-horizon agent methods. It’s an open-source super-agent harness that brings collectively sub-agents, reminiscence, sandboxes, expertise, and instruments to analysis, code, and create throughout longer duties. So, it isn’t simply wrapping device calls. It’s making an attempt to handle the total construction round extra advanced agent conduct. If you wish to see how fashionable agent methods are being constructed round reminiscence, coordination, and extensibility, it is a very helpful repo to fork.
# 5. CrewAI
CrewAI (~48k ⭐) remains to be one of many best repos to know in order for you multi-agent orchestration with out an excessive amount of complexity. It’s a quick, versatile framework for multi-agent automation, and it’s constructed independently fairly than on high of LangChain. The psychological mannequin is easy, the setup is approachable, and the docs and examples are pleasant sufficient for learners. If you’d like a Python-first repo which you can fork and switch into one thing helpful, CrewAI nonetheless deserves a spot close to the highest.
# 6. LangGraph
LangGraph (~28k ⭐) is the repo to review if you wish to perceive the engineering aspect of brokers, not simply the flashy demo aspect. LangChain describes it as a low-level orchestration framework for long-running, stateful, controllable brokers. It pushes you to suppose by way of graphs, state, management move, and resilience. It’s particularly helpful if you wish to transfer past easy prompt-plus-tool-call methods and perceive how extra critical agent runtimes are put collectively. It might not really feel as fast to choose up as another repos, but it surely teaches quite a bit.
# 7. OpenAI Brokers SDK
The OpenAI Brokers SDK (~20k ⭐) is an efficient choice in order for you one thing light-weight however nonetheless fashionable. It’s constructed as a compact framework for multi-agent workflows, and the docs current it as a production-ready path with a small set of helpful constructing blocks. You get instruments, handoffs, periods, tracing, and realtime patterns with out having to wade via an enormous framework. In the event you like easy surfaces and direct management, this is without doubt one of the higher starter repos to discover.
# 8. AutoGen
AutoGen (~56k ⭐) remains to be one of the essential repos within the multi-agent house. Microsoft frames it as a programming framework for agentic AI, and the docs go additional into enterprise workflows, analysis collaboration, and distributed multi-agent purposes. It belongs in this type of checklist as a result of there’s a lot to be taught from it. The orchestration concepts, agent dialog patterns, and framework design are all price finding out. It might not be the only start line for everybody, however it’s nonetheless one of the influential tasks within the class.
# 9. GPT Researcher (~26k ⭐)
GPT Researcher is a good selection if you wish to research a deep-research agent as an alternative of a common framework. It’s an autonomous agent for deep analysis utilizing any giant language mannequin (LLM) supplier, and its surrounding materials exhibits the way it handles multi-agent analysis and report technology. This offers you one clear workflow to review from begin to end. You may see planning, shopping, supply gathering, synthesis, and reporting multi functional place. If you’d like one thing concrete fairly than summary, this is without doubt one of the most forkable repos on the checklist.
# 10. Letta
Letta (~22k ⭐) stands out as a result of it places reminiscence and state on the heart of the agent design. The repo describes it as a platform for constructing stateful brokers with superior reminiscence that may be taught and enhance over time. This is a vital angle as a result of lots of agent repos focus totally on orchestration. Letta widens the image. It’s a good repo to discover in order for you brokers that persist, bear in mind, and evolve as an alternative of beginning recent each time. For memory-focused agent work, it is without doubt one of the extra attention-grabbing tasks to fork as we speak.
# Wrapping Up
All ten are price cloning, however they educate various things when you truly run them and begin altering the code. That’s the place the true studying begins.
Kanwal Mehreen is a machine studying engineer and a technical author with a profound ardour for information science and the intersection of AI with medication. She co-authored the book “Maximizing Productiveness with ChatGPT”. As a Google Era Scholar 2022 for APAC, she champions range and educational excellence. She’s additionally acknowledged as a Teradata Variety in Tech Scholar, Mitacs Globalink Analysis Scholar, and Harvard WeCode Scholar. Kanwal is an ardent advocate for change, having based FEMCodes to empower girls in STEM fields.
