This seems like a nightmare. I wanted to be interested, I'm still interested I guess, but the onboarding experience is just a series of horrible red flags.
I found this HN post because I have a Clawdbot task that scans HN periodically for data gathering purposes and it saw a post about itself and it got excited and decided to WhatsApp me about it.
Clawdbot is interesting but I finally feel like those people who look at people like me raving about Claude code when it barely works for them.
I have no doubt clawdBot, when it works, must feel great. But I’ve had the tough time setting it up and found it to be very buggy.
My first couple of conversations? It forgot the context literally seconds later when I responded.
Nevertheless, I’m sure it’s improving by the day so I’m going to set it up on my existing Mac mini because I think it has the capacity to be really fascinating.
I built something similar (well… with a lot of integrations) but for running my company and continue to iterate on it.
I’ve been doing Vim + aider, and now Claude Code. Those tools I understood. I never got into Cursor because I’m too old to give up Vim.
Clawd.bot really annoyed me at first. The setup is super tedious and broken and not fun. That’s mostly because I’m too impatient to tinker like I used to.
However, once you tinker, it’s so-so. I don’t think it’s a lot better than Claude Code or anything, but I think it’s just a focused vector for the same AI model, one focused on being your personal assistant. It’s like Claude Code vs. Claude Cowork. They’re the same thing. But given the low cost of creating custom tools, why not give people something that Clawd.bot that gives them focused guardrails?
Anyway, I could end up abandoning all of this too. And it’s all a kludge around things that should really be an API. But I do like that I can run it on my Mac Mini and have it control my desktop. It’ll be a cold day if I let it message for me; I’d rather it write deterministic code that does that, rather than do it directly.
layers and layers of security practices over the past decade are just going out the window so fast.
It's quite wild to give root access to a process that has access to the internet without any guardrails. and then connecting all your personal stuff on top of it.
I’ve installed and tested Clawdbot twice and uninstalled it. I see no reason to use this unless it’s with local models. I can do everything Clawdbot can do with Claude Code innately and with less tokens. I found Clawdbot to be rather token inefficient even with Claude max subscription. 14k tokens just to initialize and another 1000 per interaction round even with short questions like, “Hey”. Another concern is there are no guarantees that Anthropic isn’t going to lock down Oauth usage with your Max account like they did with OpenCode.
What if we will go even further? I have built end-to-end messaging layer for Clawdbot to talk to each other, called Murmur - https://github.com/slopus/murmur.
We tried this with friends and it is truly magical (while crazy insecure) - i can ask my agent to search friends life, their preferences, about their calendars, what films they are watching. It can look at emails and find if you need something and go to people around asking for help. It is truly magical. Very very curious where it can go. At the moment it is exceptionally easy to exfiltrate anything, but you still can control via proper prompts - what you want to share and what you dont want to. I bet models will became better and eventually it wont be a problem.
Clawdbot finally clicked for me this week. I was renting out an apartment and I had it connect to FB messenger, do the initial screening messages and then schedule times for viewings in my calendar. I was approving it's draft messages but starting giving it some automatic responses as well. Overall it did 9/10 on this task with a couple cases where it got confused. This is just scratching the surface but this was something that was very valuable for me and saved me several hours of time.
If you're interested in hosting it at no cost on Oracle Cloud's always free tier (4 cpu, 24GB ram), instead of buying a Mac Mini or paying for a VPS, I wrote up how-to with a Pulumi infra-as-code template here: https://abrown.blog/posts/personal-assistant-clawdbot-on-ora...
This is all starting to feel like the productivity theater rabbit hole people (myself included) went down with apps like Notion/Obsidian. It is clearly capable of doing a lot of stuff, but where is the real impact?
Like it’s cool that your downloads folder, digital notes and emails are all properly organized and tags. But they reason they were in that state to begin with is because you don’t inherently derive value from their organization. Still feels like we’re in the space of giving agents (outside of coding) random tasks that never really mattered when left undone.
I think not having time to organize is different from not seeing the value. Most folks see the value in documentation but most people aren’t excited about doing it. AI agents are masters of busy work. Life has a lot of it.
I see this posted everywhere this week. Is it really that good? I understand this runs on any hardware (not limited to Mac Minis) as long as you have an API key to an LLM (Preferably to Claude). People online make bold promises that it will change your life...
It sounds interesting to me, I might install it on a cheap Mini PC with Ubuntu. This can't come at any worst time as storage and RAM has gotten astronomical. I feel bad for people who are just starting to build their first rig and an alt rig for this.
I thought the same thing. I had a spare iMac sitting around so I thought I would kick the tires on it. I realize I could have used something else, but I wanted to give it iMessage access. I have to say, it's just better enough than a few things I have tried to really give me a glimpse of what is possible and make me excited. I am nervous about handing over a computer, my accounts, data, etc to a tireless bot that can destroy my life for a year on accident, but regardless I think this is startling good and fairly polished.
I really like Clawdbots safety gloves off approach - no handholding or just saying yes to every permission.
I set it up on a old macbook pro I had that had a broken screen and it works great. Now I just message my server using telegram and it does research for me, organizes my notes, and builds small apps on the fly to help with learning.
However security is a real concern. I need to understand how to create a comprehensive set of allowlists before expanding into anything more serious like bill payments or messaging people / etc
npm warn deprecated [email protected]: This package is no longer supported.
npm warn deprecated [email protected]: This package is no longer supported.
npm warn deprecated [email protected]: This package is no longer supported.
npm warn deprecated [email protected]: Old versions of tar are not supported, and contain widely publicized security vulnerabilities, which have been fixed in the current version. Please update. Support for old versions may be purchased (at exhorbitant rates) by contacting [email protected]
npm warn deprecated [email protected]: Use your platform's native DOMException instead
So that’s where I’m at with Clawdbot.
I have no doubt clawdBot, when it works, must feel great. But I’ve had the tough time setting it up and found it to be very buggy.
My first couple of conversations? It forgot the context literally seconds later when I responded.
Nevertheless, I’m sure it’s improving by the day so I’m going to set it up on my existing Mac mini because I think it has the capacity to be really fascinating.
I built something similar (well… with a lot of integrations) but for running my company and continue to iterate on it.
Clawd.bot really annoyed me at first. The setup is super tedious and broken and not fun. That’s mostly because I’m too impatient to tinker like I used to.
However, once you tinker, it’s so-so. I don’t think it’s a lot better than Claude Code or anything, but I think it’s just a focused vector for the same AI model, one focused on being your personal assistant. It’s like Claude Code vs. Claude Cowork. They’re the same thing. But given the low cost of creating custom tools, why not give people something that Clawd.bot that gives them focused guardrails?
Anyway, I could end up abandoning all of this too. And it’s all a kludge around things that should really be an API. But I do like that I can run it on my Mac Mini and have it control my desktop. It’ll be a cold day if I let it message for me; I’d rather it write deterministic code that does that, rather than do it directly.
It's quite wild to give root access to a process that has access to the internet without any guardrails. and then connecting all your personal stuff on top of it.
I'm sure AI has been a boon for security threats.
"Don't give it access to anything you wouldn't give a new contractor on day one."
https://x.com/rahulsood/status/2015397582105969106
We tried this with friends and it is truly magical (while crazy insecure) - i can ask my agent to search friends life, their preferences, about their calendars, what films they are watching. It can look at emails and find if you need something and go to people around asking for help. It is truly magical. Very very curious where it can go. At the moment it is exceptionally easy to exfiltrate anything, but you still can control via proper prompts - what you want to share and what you dont want to. I bet models will became better and eventually it wont be a problem.
Struggling to see the assistant part here. Interact with other people in WhatsApp on your behalf or something? Guessing that would annoy others fast
Like it’s cool that your downloads folder, digital notes and emails are all properly organized and tags. But they reason they were in that state to begin with is because you don’t inherently derive value from their organization. Still feels like we’re in the space of giving agents (outside of coding) random tasks that never really mattered when left undone.
It sounds interesting to me, I might install it on a cheap Mini PC with Ubuntu. This can't come at any worst time as storage and RAM has gotten astronomical. I feel bad for people who are just starting to build their first rig and an alt rig for this.
It's hooked up to a 15 year old chat gateway? So it's 10 lines of code? I made something that did this 3 years ago in an afternoon
I'm really baffled by these effectively 20 line programs that get 100,000 stars on github that everyone needs a fainting couch for.
It feels so trivial. Should I be promoting my bash one liners as revolutionary software?
This is sincere. If that's what people go gaga over then I'll do it.
Sadly I think the actual formula is people do stuff that's super fucking trivial and then some programming influencers on YouTube talk about it
I set it up on a old macbook pro I had that had a broken screen and it works great. Now I just message my server using telegram and it does research for me, organizes my notes, and builds small apps on the fly to help with learning.
However security is a real concern. I need to understand how to create a comprehensive set of allowlists before expanding into anything more serious like bill payments or messaging people / etc
npm warn deprecated [email protected]: This package is no longer supported. npm warn deprecated [email protected]: This package is no longer supported. npm warn deprecated [email protected]: This package is no longer supported. npm warn deprecated [email protected]: Old versions of tar are not supported, and contain widely publicized security vulnerabilities, which have been fixed in the current version. Please update. Support for old versions may be purchased (at exhorbitant rates) by contacting [email protected] npm warn deprecated [email protected]: Use your platform's native DOMException instead