LangManus: An Open-Source Manus Agent with LangChain + LangGraph | Hacker News Reader

LangManus: An Open-Source Manus Agent with LangChain + LangGraph

github.com

127 pointsgfortaine14 comments

Hey Hacker News, I’m excited to share LangManus, an open-source project that’s all about community-driven AI automation. Built on the shoulders of amazing open-source work, LangManus combines language models with specialized tools to tackle tasks like web search, crawling, and Python code execution—while giving back to the community that inspires it. At its core, LangManus uses a hierarchical multi-agent system, where a supervisor coordinates a team of specialized agents to handle complex workflows. It integrates with LLMs like Qwen and powers up with tools like Tavily for search and Jina for crawling. Plus, it comes with Docker support and a web UI to make interaction a breeze. Want to see it in action? Check out the demo video: https://youtu.be/sZCHqrQBUGk. In it, LangManus calculates the influence index of DeepSeek R1 on HuggingFace using a fully automated plan—think web searches, data retrieval, and Python code crunching, all seamlessly tied together. Here’s what makes LangManus stand out: Multi-Agent Magic: A supervisor delegates tasks to agents like Researcher, Coder, and Browser for efficient execution.

Tool Integration: Web search with Tavily, neural search with Jina, and Python REPL for coding on the fly.

Ease of Use: Docker-ready and a web UI for quick setup and control.

If you’re into AI automation, multi-agent systems, or just love contributing to open-source projects, swing by the GitHub repo: https://github.com/langmanus/langmanus. Dive into the code, play with the demo, or drop some feedback—we’d love to hear from you. Join the community and let’s build something awesome together!

How do you think about core business logic (or at least, _significant_ business logic) being embedded within prompts like here: https://github.com/langmanus/langmanus/blob/main/src/prompts...

Do you think or worry about not-being able to test these things? (Or is that just me :))

Details: I ack/understand this comes from a dependency (ReAct agents); not directly langmanus.

But, still, curious what the community/hn-tech thinks of testability, veracity, potentially conflicting or overlapping instructions across agents, etc, wrt “prompts” as sources of logic. Ack its a general practice with LLMs.

Thought the logic was going to go with cpu and inference related tasks will use GPU.

Super cool. Congrats on the launch!

Let me know if we can help with something at Hyperbrowser.

Since "Manus" is a registered trademark, you should probably change the name as quickly as possible to avoid legal issues with the original trademark holder. At the very least, you could remove the "M" to avoid directly copying it while still maintaining recognizability.

I see what you did there

You should probably change the name, I thought it was associated with Manus usually projects like this will specifically have in the tag line “An Open Source alternative to X.”

Other than that looks very cool, reading through the code now. A lot of these projects very heavily lean on Browser Use.

Branding and UX are very under appreciated in the OS community.

Within a week this was made unavailable, did you choose to rename the project? You might want to leave a notice of where to find the new repository.

Would be great if you put together a start-to-finish tutorial showing the step-by-step process from initial Git clone to completed successful run.

Being able to see the prompts is incredibly useful - thanks so much for this - great resource.

Congrats on the launch!

Very cool.

cool

Reply on news.ycombinator.com