5 ms·
Honest question: What do people actually USE OpenClaw for? The most common usage seems to be "it reads your emails!", that's the exact opposite of "exciting"...
by sunaookami 7mo ago
Honest question: What do people actually USE OpenClaw for? The most common usage seems to be "it reads your emails!", that's the exact opposite of "exciting"...
- browningstreet 7mo ago[flagged]
- freedomben 7mo agoyeah I don't normally say "read previous HN articles" but it has been asked at least once in every article here.
- sunaookami 7mo agoObviously I already searched the web (not specifically HN I must admit) and there were always incredibly generic non-answers that ultimately say nothing (and they assume you have 3000$ per month or 2000 Mac Minis on your desk (hyperbole)).
- ziml77 7mo agoIncredibly, one of the responses you got already is exactly one of those replies that says nothing. There's a whole bunch of words that don't actually answer the question.
- emp17344 7mo agoI think you’ve got your answer, then. If nobody can tell you what it’s really used for, it likely doesn’t have any real use cases.
- _doctor_love 7mo agoAssuming you're asking in good faith, IMHO the deeper story around OpenClaw is that it's the core piece of a larger pattern. The way I'm seeing folks responsibly use OpenClaw is to install it as a well-regulated governor driving other agents and other tools. It is effectively the big brain orchestrating a larger system. So for instance, you could have an OpenClaw jail where you-the-human talk to OpenClaw via some channel, and then that directs OpenClaw to put lower-level agents to work. In some sense it's a bit like Dwarf Fortress or the old Dungeon Keeper game. You declare what you want to have happen and then the imps run off and do it. [EDIT: I truly down understand sometimes why people downvote things. If you don't like what I'm saying, at least reply with some kind of argument.]
- mvdtnz 7mo agoYou're probably being downvoted because you didn't answer the question. The questioner specifically asked what people are using it for and you answered by describing your technical setup. What we want to know is, what are you actually achieving with this tool?
- j-bos 7mo agoSo I neither downvoted nor upvoted you, but I think people may be downvoting, in addition to the fact that they just don't like the thing, based on the fact that you didn't directly answer the question. Specifically, what are you using it for, not what hypothetically it would be used for.
- PKop 7mo agoFirst words out of your mouth are to accuse OP of not seriously asking the question. Then you write paragraphs saying nothing much at all. You could have simply answered the question in a simple straightforward manner.
- _doctor_love 7mo agoMan, all the replies to my comment. Do you guys know how to fucking read?
- PKop 7mo agoName 2 things you actually do with OpenClaw. And don't swear in your response.
- the_pwner224 7mo agoYou have yet to answer the original question - what do you actually do with OpenClaw? A concrete example of something that actually happens, not a system architecture description.
- bongripper 7mo ago[dead]
- sgillen 7mo agoI've only been playing with it recently ... I have mine scraping for SF city meetings that I can attend and public comment to advocate for more housing etc (https://github.com/sgillen/sf-civic-digest https://github.com/sgillen/sf-civic-digest). It also have mine automatically grabs a spot at my gym when spots are released because I always forget. I'm just playing with it, it's been fun! It's all on a VM in the cloud and I assume it could get pwned at any time but the blast radius would be small.
- mvdtnz 7mo ago[flagged]
- gruez 7mo agoNo? The comment was admittedly ambiguous but if you go to repo it's far clearer: >I use it to give me a weekly digest of what happened in my neighborhood and if there are any public hearings or trash pickups I might want to attend.
- WhrRTheBaboons 7mo agothat does not seem like something you need an 'autonomous' agent for.
- Sohcahtoa82 7mo agoWhat would you propose as an alternative? Anything not relying on an LLM likely means having to write bespoke scripts. That's not really worth the time, especially when you want summaries and not having to skim things yourself. Going from doing it manually on a regular basis to an autonomous agent turns a frequent 5-15 minute task into a 30 second one.
- mvdtnz 7mo ago> Anything not relying on an LLM likely means having to write bespoke scripts. The very first line in your readme is "CivicClaw is a set of scripts and prompts" though? And almost the entire repo is a bunch of python scripts under a /scripts folder. I looked at one randomly chosen script (scripts/sf_rec_park.py) and it's 549 lines of Python to fetch and summarise data that is available on an RSS feed ( https://sanfrancisco.granicus.com/ViewPublisher.php?view_id=91 https://sanfrancisco.granicus.com/ViewPublisher.php?view_id=... )
- dyauspitr 7mo agoAgent based chron jobs mostly that work with other agents. It’s really nice if you want to tell your computer to do something repeatedly or in confluence with many other agents in a very simple way. Like check my email for messages from Nadia and send me a notification and turn on all the lights in my driveway when she gets there without having to actually get into the nuts and bolts of implementing it. It’s actually really powerful and probably what Siri should be.
- rustystump 7mo agoI think this is close to the head of the nail. It kinda unlocks handling novelish asks that previous siri/alexa just couldnt handle. As long as a thing has well documented api spec then it instantly is usable. This makes the clawbot flow extraordinarily more useful. I think devs are too focused on the technical what did u build with it. For example. My brother runs a small recruiting agency. Super nontechnical. Out of nowhere he asks me about openclaw. Then with no help, he sets it up and uses it. Still no help, he has all kinds of nonsense hooked up and running blowing through tokens. He is blown away by it and wants to get it for all of his employees. He thinks about it in terms of cost per min running and not in tokens. This is the sticky gooey value to whatever openclaw is doing.
- earnesti 7mo agoI use it for a side project. I just put it on VPS, and then it edits the code and tests it. The nice thing is that I can use it on the go whenever I have spare moment. It is addictive, but way better addiction than social media IMO. The thing where you give it access to all your personal data and whatever I haven't done and wouldn't do.
- throwatdem12311 7mo agoI can do this with Claude Code from my phone. What’s the difference?
- franze 7mo agomy claw controls my old M2 mac, mostly my claw uses Claude code to code
- operatingthetan 7mo agoSo you're using a different llm to control claude code to get around the Anthropic TOS about openclaw usage?
- paganel 7mo agoAt this point I'm personally lost, unless GP's comment wasn't some sort of satire (which would be valid, this being a topic about AI).
- emptysongglass 7mo ago[flagged]
- DonHopkins 7mo agoBefore I decide to shoot up smack, I like to ask junkies what the whole heroin experience is like, what they use it for, and how it has affected their lives. Nina Hagen - Smack Jack https://www.youtube.com/watch?v=nIDnN34ZZaE https://www.youtube.com/watch?v=nIDnN34ZZaE >Smack Ist Dreck, Stop It Oder Verreck!
- emptysongglass 7mo agoSo you're comparing a generic tool you can tailor to your own needs to drugs? This is exactly why I have zero interest in engaging with people over this topic.
- DonHopkins 7mo ago[flagged]
- equasar 7mo ago[flagged]
- emptysongglass 7mo agoWhy don't you try it yourself instead of making uninformed claims
- DonHopkins 7mo ago[flagged]
- equasar 7mo agoWhy would I do that? I am entirely good using LLMs like Claude building tools for me. There's no use case for OpenClawthat I am aware of can replace of what I have/need. I think it makes my point strong, people who uses OpenClaw, might be lazy on how to do things properly with LLMs.
- knights_gambit 7mo agoI use it to manage a media server. And use natural language to download movies and series. Also I use to for homeassistant so I csn use natural language for vacuuming the house and things like that. I do use it for a number of other tasks but those are the most partical.
- nickthegreek 7mo agoGood use cases, but I do want to point out that you can do all of that with HA itself. Are you using skills to talk to *arr services?
- FrameworkFred 7mo agoso far, I've used it to kill a bunch of time trying to get it to respond to "Hi @Kirk" in a private Slack channel. ...and to laugh a little every time it calls me "commander" or asks "What's the next mission?" or (and this is the best one) it uses the catchphrase I gave it which is "it's probably fine" (and it uses it entirely appropriately...I think there must have been a lot of sarcasm in qwen 3.5's training data) and I've treated it like it's already been compromised the whole time.
- globular-toast 7mo agoSo basically an eggdrop like we had in the 90s except, by the sounds of it, less useful and considerably less fun.
- nickthegreek 7mo agoHaving this in a discord is actually like having an eggdrop on steroids. I would of lost my mind having this on efnet in the late 90s.
- rubslopes 7mo agoI don't use this one, but a simpler one, also running on a vps. I communicate via telegram. I say to it: check my pending tasks on Todoist and see if you can tackle on of those by yourself. It then finds some bugs in a webapp that I took note. I tell it to go for it, but use a new branch and deploy it on a new url. So it clones the repo, fix it, commit, push, deploy, and test. It just messages me afterwards. This is possible because it has access to my todoist and github and several other services.
- operatingthetan 7mo agoI use it mostly for the crons, it runs a personal productivity system that tracks my tasks, provides nudges, talks through stuff etc. It's all stored in an Obsidian vault that syncs to my desktop. I don't use it to control email/calendars or other agents.
- veganmosfet 7mo agoI am experimenting prompt injection on OpenClaw [0][1], quite exciting. [0] https://itmeetsot.eu/posts/2026-03-27-openclaw_webfetch/ https://itmeetsot.eu/posts/2026-03-27-openclaw_webfetch/ [1] https://itmeetsot.eu/posts/2026-03-03-openclaw3/ https://itmeetsot.eu/posts/2026-03-03-openclaw3/
- sunaookami 7mo agoAwesome and very interesting posts, thanks for sharing! Always reminds me of the "lethal trifecta": https://simonwillison.net/2025/Jun/16/the-lethal-trifecta/ https://simonwillison.net/2025/Jun/16/the-lethal-trifecta/
- veganmosfet 7mo agoYou're welcome! My main takeaway message is: models (even opus4.6) do not follow security "instructions" reliably. In OpenClaw, they added security warnings, tags, random IDs... None of these countermeasures work reliably. Even sandboxing can be escaped (not in the classical sense using vulnerabilities, but using multi-layered prompt injection payload with natural language only)[0]. As soon as untrusted content is injected in the context, do not trust any actions downstream. [0] https://itmeetsot.eu/posts/2026-02-15-openclaw_sandbox/ https://itmeetsot.eu/posts/2026-02-15-openclaw_sandbox/
- cornholio 7mo agoWhat do you think about CaMeL and similar approaches? https://simonwillison.net/2025/Apr/11/camel/ https://simonwillison.net/2025/Apr/11/camel/
- veganmosfet 7mo agoGood question. CaMeL is imho safer, but hard to implement into modern agents like OpenClaw. Its core idea is that a privileged LLM plans from the (trusted) user request only, while a restricted interpreter executes that plan (and enforces policies). Untrusted content is parsed separately and is not fed back into the privileged LLM. Modern agents are useful exactly because they run a feedback loop (observe, reason, adapt, use tools, repeat). CaMeL breaks that loop, which improves security but makes it a poor fit for highly general agents like OpenClaw.
- qingcharles 7mo agoI was asked by someone recently to try to set up an OpenClaw that would search for ordinances and other land registry information for all 3000+ counties/parishes in the USA to obtain and distill specific details on their support for building tiny homes.
- mvdtnz 7mo agoWhat is OpenClaw doing here that Claude Desktop or Claude Code couldn't do?
- qingcharles 7mo agoClaude Desktop and Code are built for synchronous, human-in-the-loop interactions. Scraping 3000 janky municipal websites, you need a "fire-and-forget" background worker. Claw lets you kick off a massive job and just get a ping when it's done. I'd also instantly hit Claude Desktop's rate limits with this I reckon. Since Claw uses APIs, you bypass those limits and can route the messy scraping to cheap models, saving expensive ones for the actual analysis. It also handles Playwright integration and state persistence out of the box so a crash doesn't wipe your progress. If I'm wrong, I'm open to learning. I'm as new to this as everyone :)
- dgb23 7mo agoI would first automate everything with scripts, and only use an agent for the parts that require it. For example you mentioned playwright? That can be automated. It doesn’t need to be a free form tool that the agent uses at will. If that means the scripts need to be adopted to changes, then that’s a separate, controlled workflow. This approach can save you a ton of tokens, increasee reliability and observability, and it saves compute as well. Sometimes it‘s useful to let the agent do things fully agentic, so you can then iteratively extract the deterministic parts.
- RIMR 7mo agoI give it monumental tasks. For example, I will write massive markdown files describing all the features I want to see in an application, and I will use a standard AI chatbot to check my work and consider additional details. Finally, when I have everything written down, I upload it to OpenClaw and tell the agent to make it happen. Sometimes it toils away for 2+ hours, spawning Claude Code instances, checking its work, testing the code, even using browser automation to make sure everything works the way it is supposed to if it's writing a webapp. In the end, it consumes like $10-20 worth of tokens and spits out a functional application with everything I asked for. Claude Code can do this on its own, to an extent, but there's something about getting OpenClaw to iterate through multiple sessions and testing everything to make sure it works the way I described that I really like. It completely offloads the process to the AI, and keeps me mostly out of the loop. Is the code any good? Probably not. Am I at risk of being exploited by malware? Probably. But I have automated quite a lot of things with the software that OpenClaw builds for me, and I am careful to review the libraries it imports before running the code on any machine with actual access to anything I actually care about. Personally, anyone using OpenClaw for the "it reads your emails" use case is crazy, because prompt injection is real, and you're basically inviting anyone who knows your email address to take a stab at pwning you, with full access to your personal life. I keep my instances on a VPS, behind a restrictive security group, and only accessible via Tailscale where it has zero access to anything on my tailnet. I only recently gave it its own email account (not mine!), but even then I am skeptical of doing so, and take efforts to prevent it from taking action on any email it receives (e.g., disabling the Heartbeat) because who knows what it'll end up doing. I mostly like that it can email me if I ask it to.