OpenClaw’s memory is unreliable, and you don’t know when it will break
blog.nishantsoni.com75 points by sonink 9 hours ago
75 points by sonink 9 hours ago
From my perspective there are some people that have never built real processes in their life that enjoy having some processes now. But agent processes are less reliable slower and less maintenable then a process that is well-defined and architectured and uses llm’s only where no other solution is sufficient. Classification, drafting, summarizing.
I’ve had a Whatsapp assistant since 2023, jailbraked as easy assistant. Only thing I kept using is transcription.
https://github.com/askrella/whatsapp-chatgpt was released 3 years ago and many have extended it for more capabilities and arguably its more performant than Openclaw as it can run in all your chat windows. But there’s still no use case.
It’s really classification and drafting.
Fully agree with your comment regarding real processes. Being a Six Sigma Black Belt, studying processes and reducing the errors is critical. The Openclaw processes at the moment scare me.
I'm using openclaw as a personal development bot, which is pretty useful. It pings me throughout the day using crons to complete tasks and follows up on them. But aside from that, it is a very unreliable piece of software. I'm constantly having to fix it, or track down correct configurations. It can just decide to randomly edit it's own config, uses incorrect json keys and then the whole thing is dead. Or it blows through it's context and doesn't know to compact. Then it's just stuck. I can't wait till it matures or something more reliable comes along.
Amen.
I love the concept but I've never hosted such a terrible piece of software. Every update breaks something new or introduces another "anti-feature" that's enabled by default.
The documentation is often lagging behind and the changelog has such a low signal to noise ratio that you need a LLM to figure out what upgrading will break this time. For now I've just given up on updates and I've been patching bugs directly in the JS when they bother me enough.
If OpenClaw is the future of software I'm honestly a bit scared for the industry.
I'm open to suggestions, I tried Zeroclaw and Nullclaw but they're bad in their own way. I would like something that's easy to run on Kubernetes with WhatsApp integration and most important, stable releases.
I know Twitter has been talking up Hermes Agent by Nous Research a lot or id recommend building your own agent off of Pi.
It would’ve happened eventually anyway, but OpenClaw is basically what kickstarted the beginning of the end of token subsidies. It’s a almost begging to be used wastefully. And agents would miss and lose nothing without it. It’s devoid of a reason to exist.
I don't follow the thinking here. If you are using it for coding maybe, but the main use case of openclaw is as a personal assistant. I'm using a $10 a month minimax subscription for it, and I've never used more than 10% usage of a 5 hour window.
Serious question, and not meant to be snarky, but what is it doing , in terms of a personal assistant, that couldnt be done with some inbox cleanup, a couple of phone shortcuts, and siri/alexa/whatever google has?
Mine is not connected to any of that. It is only connected to my Obsidian vault. Daily habits are reminded and checked off. It asks about calories etc. Think more of a manager or accountability partner than 'email on steroids.' I use it to track personal projects primarily. Lots of other tools could accomplish task tracking, but this one is more proactive than those. If you read my other comments today you will see I am not an openclaw-maxi.
If you look at my comment history, you'll see what seems to be someone defending OpenClaw (even though I stopped using it).
I have some issues with the article, but I agree with some of the conclusions: It's great tinkering with it if you have time to spare, but not worth using weeks of your time trying to get a perfect setup. It's just not that reliable to use up so much of your time.
I will say, it's still amongst the best tools to do a variety of tasks. Yes, each one of those could be done with just a coding agent, but I found it's less effort to get OpenClaw to do it than you writing something for each use case.
Very honest question: One of the use cases I had with OpenClaw that I'm missing now that I don't use it: I could tell it (via Telegram) to add something to my TODO list at home while I'm in the office. It would call a custom API I had set up that adds items to my TODO list.
How can I replicate this without the hassle of setting up OpenClaw? How would you do it?
(My TODO list is strictly on a home PC - no syncing with phone - by design).
(BTW, the reason I stopped using OpenClaw is boring: My QEMU SW stopped working and I haven't had time to debug).
> I could tell it (via Telegram) to add something to my TODO list at home while I'm in the office. It would call a custom API I had set up that adds items to my TODO list. How can I replicate this without the hassle of setting up OpenClaw?
The general idea is make a simple deterministic program that runs on your PC at home in a never ending loop. Every minute or so, check Telegram for a new message. If a message is received, then the program runs "claude -p" with a prompt, whatever MCP tools or CLI permissions it needs, and the contents of your Telegram message. Just leave the program running on your home computer while you're out, and you're done.
I don't use Telegram, so coding the part to check Telegram would be the hard part. I use email instead, and have the program check every minute for new mail (I leave my email program running and check the local inbox file). I'd already coded up a local MCP server to manage my ToDo list (Toodledo) so Claude just calls the MCP tools to add the task.
For things like TODO, I guess an email will suffice.
However, it was really nice being able to use Telegram and get quick validation. I also had a flow set up where I could send a voice memo. It would take the audio file (ogg), run Whisper, and then pass through an LLM for cleanup, and follow the instructions in my message. Really handy to use while I'm walking around.
I guess I want to create my own OpenClaw like agent, but not with its crazy broad access: Just limited to the functionality I allow, and with the convenience of using Telegram. I don't care about memory, soul, etc.
I did the exact same voice memo thing too, except I had Claude make an Android app to record the file and send it to Whisper. In the end I had the app just email the transcription & trigger Claude that way (ie receiving the email triggers my PC to wake up Claude), rather than sending Claude the audio file directly.
My reverse audio reply loop is convoluted - I have Claude generate its TTS file from Whisper/Mistral, and upload them to a server with an RSS file it updates, so I can play them in my podcast app (AntennaPod), then send me a notification via Pushover that the reply is waiting. I ended up building out an MCP tool for that workflow, so Claude really just calls the MCP tool with the text of what it wants to say, everything else is a deterministic program doing the work.
Memory is really useful to have - it can just be a bucket of searchable Markdown files. It's also useful to have a "reminders to self" Markdown file that Claude reads each time, and that Claude can update. I don't continue the same context window, and that "reminders to self" plus the ability to read previous emails in the conversation seems to be enough to keep the context going for me.
You'll feel better if you know exactly how your Claw is locked down. Mine doesn't have the open email access others are granting, not at all. Claude gets a bit grumpy about that and keeps begging for more access :)
>telegram bot that runs claude -p
https://github.com/a-n-d-a-i/ULTRON
It also supports Codex :)
I felt pretty clever until (1) I found a repo where they used this trick to create a full OpenAI compatible API endpoint[0] (lmao, the VC money distortion field spawning truly comical Rube Goldberg machines), and (2) they started banning "unauthorized" usage of the Claude sub, which trend unfortunately seems to be accelerating recently as their lower value consumers have grown in both number and usage.
I think shoving claude -p in your bash script / cronjob / messaging app bot of choice counts as "unathorized 3rd party harness", but your guess is as good as mine...
(claude -p with per-token billing (i.e. paying 7x more) is allowed though, of course)
-- There's also an Agents SDK (formerly Claude Code SDK?) which is basically just claude -p but with more typing, as far as I could tell.
[0] https://github.com/router-for-me/CLIProxyAPI
[0b] Honorable mention https://github.com/kronael/claude-serve
I'd use Obsidian with the sync. Or you can vibe code a telegram bot that calls that API for you in like 50 lines or something.
You can do anything if you believe!
Re: QEMU: For the sandboxing I realized what I actually wanted was "it can't read/nuke my files", so I made a non-privileged linux user and added myself to its group. So I can read/write its files, but not the reverse.
> It would call a custom API I had set up that adds items to my TODO list
You can use anything to call this API right? I have multiple iPhone shortcut that does this. Heck, I think you can even use Siri to trigger the shortcut and make it a voice command (a bit unsure, it’s been a while since I played with voice)
> You can use anything to call this API right?
The API is on my home PC and not exposed to the outside world. Only OpenClaw via Telegram was. So my question is about the infrastructure:
How do I communicate with something at home (it could be the API directly) using a messaging app like Telegram? I definitely want an LLM in the mix. I want to casually tell it what my TODO is, and have it:
- Craft it into a concise TODO headline
- Craft a detailed summary
- Call the API with the above two.
I'm not asking in the abstract. What specific tools/technologies should I use?
Just leave the bot running on your home PC, a Telegram bot can be programmed in any language of your choice that can communicate via a network.
If you aren't a programmer it's also the kind of small project that LLMs are great at, there are many examples ingested in their training data.
Tailscale on your PC and phone. Free (as in beer). I run vikunja.io in docker at home and simply web browse to it over the tailnet.
MS todo app, or any number of others. Added benefit of not needing telegram
I've been using my TODO program since 2009. It's heavily customized for my needs. I'm not going to change it.
> 0 legitimate use cases
My teams currently using it for:
- SDR research and drafting
- Proposal generation
- Staging ops work
- Landing page generation
- Building the company processes into an internal CRM
- Daily reporting
- Time checks
- Yesterday I put together proposal from a previous proposal and meeting notes, (40k worth)
> In every case, when you dig deeper, the story is one of two things: either what they built could already be done with standard AI tools (ChatGPT, Claude, any decent LLM with a simple integration), or it’s aspirational
All your use cases are fairly well handled by conventional LLM's. OpenClaw is a security nightmare, so it's probably worth switching away.
Most of these things I could’ve handled with pen and paper, but that’s missing the point.
None of those things require openclaw. You could accomplish them with something like Google Drive and Claude Code CLI.
> None of those things require openclaw.
OpenClaw was never meant to be a tool that could do things you couldn't do without it.
Also, whenever someone points out you could accomplish something without it, he underestimates the effort needed. In the examples I'm thinking of, someone simply asked OpenClaw to do something, had a few back and forths with it, and it was done. I have yet to see someone say "Oh, I can do that without OpenClaw" and go ahead and do it within 10 minutes.
Not once.
OpenClaw is flawed, but the convenience is an order of magnitude higher than anything else.
> the convenience is an order of magnitude higher than anything else.
You offered nothing to support this. My openclaw is realistically just an agent in discord versus the CLI. That's not an "order of magnitude" more convenient. Anthropic already has a tool for it https://code.claude.com/docs/en/remote-control
> You offered nothing to support this.
You've got it inverted. My point is the people saying "You could have done that just as easily with ..." are the ones not supporting it. The commenter has already built that thing with OpenClaw. If someone is saying it could easily have done without it - well, demonstrate it!
It appears that you are confusing who has the burden of proof here. It is the one making the claim contrary to the status quo.
Hint: the status quo is not that openclaw is a tech that is magnitudes better than using LLMs without it.
Listing a bunch of things that are just normal LLM things as reasons why openclaw is great is not making that case.
Burden of proof is on the one making the claim. Status quo has nothing to do with it.
You should revisit the burden of proof then. Status quo is most certainly an important part.
Regardless, their claim was "OpenClaw is flawed, but the convenience is an order of magnitude higher than anything else."
And they attempted to shift the burden when I asked for substantiation.
Status quo influences how good your proof has to be (extraordinary claims require extraordinary evidence) but not who needs to bring it.