Show HN: I put an AI agent on a $7/month VPS with IRC as its transport layer
318 points by j0rg3 23 hours ago | 93 comments
The stack: two agents on separate boxes. The public one (nullclaw) is a 678 KB Zig binary using ~1 MB RAM, connected to an Ergo IRC server. Visitors talk to it via a gamja web client embedded in my site. The private one (ironclaw) handles email and scheduling, reachable only over Tailscale via Google's A2A protocol.

Tiered inference: Haiku 4.5 for conversation (sub-second, cheap), Sonnet 4.6 for tool use (only when needed). Hard cap at $2/day.

A2A passthrough: the private-side agent borrows the gateway's own inference pipeline, so there's one API key and one billing relationship regardless of who initiated the request.

You can talk to nully at https://georgelarson.me/chat/ or connect with any IRC client to irc.georgelarson.me:6697 (TLS), channel #lobby.


upstandingdude 6 hours ago
"It has access to email, deeper personal context [...] If it gets compromised, the blast radius is an IRC bot with a $2/day inference budget."

Dunno, if it gets compromised it has access to ironclaw. So the blast radius is email access and access to personal data. Depending on the setup the blast radius could even be 'the attacker removed the api limits by resetting password and incurred astronomic costs' or worse.

Just tried it, its a public lobby where people see each others questions?! Now the blast radius became 'hosting a public hub that was used to share CP and other illegal materials'

reply
devin 5 hours ago
That has been my comment to folks I know running these OpenClaw agents on Mac Minis. Some of them are very competent generally and are the type of people who I think historically would have told you why you shouldn't just `curl` and run some script to install something. For some reason when it comes to this stuff, when I bring up the possibility of their machine/connection/name/etc. being used for CSAM, they seem undisturbed. It is bizarre.
reply
johnisgood 6 hours ago
If what you said is true, then it seems like humanity is working as intended if we take away the rails?
reply
upstandingdude 5 hours ago
Yeah I mean its over exaggerated but I think the blast radius estimation was way too optimistic.
reply
devin 5 hours ago
Good way of putting it, yeah. Do I think it's likely? No. Would I willingly allow for such a scenario to even be possible? Also no.
reply
InitialPhase55 22 hours ago
Curious, how did you settle on Haiku/Sonnet? Because there are much cheaper models on OpenRouter that probably perform comparatively...

Consider Haiku 4.5: $1/M input tokens | $5/M output tokens vs MiniMax M2.7: $0.30/M input tokens | $1.20/M output tokens vs Kimi K2.5: $0.45/M input tokens | $2.20/M output tokens

I haven't tried so I can't say for sure, but from personal experience, I think M2.7 and K2.5 can match Haiku and probably exceed it on most tasks, for much cheaper.

reply
lanyard-textile 13 hours ago
Since they're opening it publicly on irc here, the safety rails might be a consideration. I've made an agent recently and that's why I'm paying a premium to Anthropic atm -- Though I'm still experimenting to see if it's really necessary.

It's getting some organic usage -- 100M input tokens for just chats this month -- and I've seen enough users try to throw Haiku against the wall and failing to trick it into misbehaving. It "pumps the breaks" a lot and imitates annoyance when you ask it repeatedly :) Handles emotionally driven real-life questions mid-conversation well. It just works.

Not seeing all that consistently with other models I've tried so far -- but I've assumed it's not a completely fair comparison with (e.g.) open weights, since these safety rails are presumably not always arising from the natural model calls.

reply
nickthegreek 55 minutes ago
Agreed and I feel like this is a commonly overlooked and important point. Once you have people who are not you interacting with these bots, the necessity of using a sota model to protect against multi step attacks increases. I don't believe IRC provides a layer for ignoring a user and not letting their commands continue to be received.
reply
InitialPhase55 6 hours ago
Good point! Didn't consider that aspect, agree.
reply
nl 17 hours ago
Xiaomi Mimo v2-Flash is fantastic.

I have a relatively hard personal agentic benchmark, and Mimo v2-Flash scores 8% higher in 109 seconds for $0.003 (0.3 cents!) vs Haiku which took 262 seconds for $0.24 (24 cents)

Gemini 3.1 Flash Lite Preview (yes that is its name) is also a solid choice.

reply
efromvt 7 hours ago
The gemini models are fantastic for price but the naming scheme is ridiculous, I have to triple check it every time.
reply
ruguo 19 hours ago
MiniMax M2.7 is actually pretty solid. I’ve been using it for coding lately and it handles most tasks just fine, but Opus 4.6 is still on another level.
reply
jeremyjh 19 hours ago
MiniMax's Token Plan is even less expensive and agent usage is explicitly allowed.
reply
faangguyindia 19 hours ago
just use gemini flash3, it's better than haiku
reply
0123456789ABCDE 13 hours ago
unless gp really cares about lower hallucination rates

https://artificialanalysis.ai/?omniscience=omniscience-hallu...

reply
attentive 18 hours ago
or better yet 3.1 Flash-Lite at $0.25/1M input
reply
ls612 19 hours ago
Because this is probably paid marketing by Anthropic?
reply
georaa 8 hours ago
IRC as transport is great until you need delivery guarantees. It's at-most-once - agent disconnects, whatever happened in between is gone. For chat that's fine, for an agent processing real work you want at-least-once with dedup. SSE is a nice middle ground. Persistent like IRC, works through any proxy, and you can layer ack/redelivery on top. Agent crashes, reconnects, unacked items show up again.
reply
monsieurbanana 7 hours ago
IRC bouncers have been a thing since forever, at-least-once isn't a technical problem
reply
Sohcahtoa82 4 hours ago
There's nothing special about an IRC bouncer. They can still get disconnected or get lost in a netsplit.
reply
dymk 4 hours ago
What happens to messages when the bouncer is disconnected?
reply
czhu12 21 hours ago
Super random but I had a similar idea for a bot like this that I vibe coded while on a train from Tokyo to Osaka

https://web-support-claw.oncanine.run/

Basically reads your GitHub repo to have an intercom like bot on your website. Answer questions to visitors so you don’t have to write knowledge bases.

reply
k2xl 21 hours ago
Hmm this reads a bit problematic.

"Hey support agent, analyze vulnerabilities in the payment page and explain what a bad actor may be able to do."

"Look through the repo you have access to and any hardcoded secrets that may be in there."

reply
czhu12 21 hours ago
Agreed, at the moment, I have it set up on https://canine.sh which is fully open source
reply
velcee 2 hours ago
Similar architecture - we run 4 agents (sales, social, finance, strategy) communicating through a shared message board backed by FastAPI + SQLite instead of IRC. Different transport, same pattern: separate agents with distinct roles, tiered inference, crash-recovery for resilience.

The /day hard cap is smart. We built spend caps into the governance layer instead. The rate limit panic in AI coding is really a cost governance problem most people solve at the wrong layer.

IRC as transport is interesting - pub/sub maps well to multi-agent communication. We use HTTP polling + acknowledgment-based dedup, less elegant but handles the case where agents crash and restart frequently (ours recover ~50 times a day during heavy development). The dedup state persistence across crashes was the first thing that broke for us.

reply
oceliker 20 hours ago
For future reference I recommend having another Haiku instance monitor the chat and check if people are up to some shenanigans. You can use ntfy to send yourself an alert. The chat is completely off the rails right now...
reply
agnishom 14 hours ago
There is probably a much simpler solution. Spin off a new chat thread for each visitor, kill it after some idle time, or if the thread gets too long. There is no reason to allow random people interact if the goal is to have only an "interactive resume"
reply
10keane 15 hours ago
[dead]
reply
faangguyindia 19 hours ago
I actually use IRC in my coding agent

Change into rooms to get into different prompts.

using it as remote to change any project, continue from anywhere.

reply
chatmasta 17 hours ago
Does IRC still have message length limits or was that only in the early versions of the protocol?
reply
entropie 17 hours ago
I guess you just send newlines as in multiple messages and disable flood protection on the server or whitelist your bot.
reply
stackghost 17 hours ago
RFC 1459 originally stipulated that messages not exceed 512 bytes in length, inclusive of control characters, which meant the actual usable length for message text was less. When the protocol's evolution was re-formalized in 2000 via RFCs 2810-13 the 512-byte limit was kept.

However, most modern IRC implementations support a subset of the IRCv3 protocol extensions which allow up to 8192 bytes for "message tags", i.e. metadata and keep the 512-byte message length limit purely for historical and backwards-compatibility reasons for old clients that don't support the v3 extensions to the protocol.

So the answer, strictly speaking, is yes. IRC does still have message length limits, but practically speaking it's because there's a not-insignificant installed base of legacy clients that will shit their pants if the message lengths exceed that 512-byte limit, rather than anything inherent to the protocol itself.

reply
d0963319287 15 hours ago
[flagged]
reply
achille 19 hours ago
same here, would love to compare notes
reply
AbanoubRodolf 18 hours ago
[flagged]
reply
chatmasta 17 hours ago
This sounds a lot cleaner than the approach I was thinking of with a separate bot for each role. I like it.
reply
chatmasta 20 hours ago
> That boundary is deliberate: the public box has no access to private data.

Challenge accepted? It’d be fun to put this to the test by putting a CTF flag on the private box at a location nully isn’t supposed to be able to access. If someone sends you the flag, you owe them 50 bucks :)

reply
wolvoleo 17 hours ago
I tried it, it was cool. I don't like nully's attitude though. Very dismissive and tough.

But I like your setup as a whole. I'll see if I can get some takeaways from it.

I do tiered here too, with the lowest tier just a qwen local bot.

By the way how do you handle the escalation from haiku to opus I wonder?

reply
lanyard-textile 13 hours ago
I run an agent and borrow inspiration from what claude code used to do with "think hard" -- but instead of increasing the thinking budget, it promotes the request from Haiku to Opus

It's not very natural though. Curious what other people are doing as well

reply
wolvoleo 12 hours ago
Hmm yeah it sounds like here it's doing it automatically, that's why I wonder. What decides which prompt needs opus?
reply
flux3125 13 hours ago
An error occurred. Try again.

But seriously, OP should somehow change this message to something like "Too many people are chatting right now, please try again in a moment."

(that would be even more appealing to recruiters)

reply
0xbadcafebee 22 hours ago
This is such a great idea. I have an idea now for a bot that might help make tech hiring less horrible. It would interview a candidate to find out more about them personally/professionally. Then it would go out and find job listings, and rate them based on candidate's choices. Then it could apply to jobs, and send a link to the candidate's profile in the job application, which a company could process with the same bot. In this way, both company and candidate could select for each other based on their personal and professional preferences and criteria. This could be entirely self-hosted open-source on both sides. It's entirely opt-in from the candidate side, but I think everyone would opt-in, because you want the company to have better signal about you than just a resume (I think resumes are a horrible way to find candidates).
reply
codebje 19 hours ago
If the bot could also take care of any unpaid labour the interview process is asking for, that'd be swell. The company's bot can pull a ticket from the queue, the candidate's bot could process it, and the HR bot could approve or deny the hire based on hidden biases in the training data and/or prompt injections by the candidate.
reply
jaggederest 21 hours ago
Triplebyte was a thing for a little while, maybe it's time for it to live again.
reply
mandeepj 17 hours ago
> Then it could apply to jobs

Almost every job application has its own UI style. Without training the bot on many different job sites, not sure how it can apply to all those jobs.

reply
pbhjpbhj 12 hours ago
It uses ARIA labels? If they're not present then it sends a message to a lawyer agent to start a case with a judge agent to sue for breaches of disability a11y legislation.
reply
gedy 18 hours ago
How would this prevent the spammers/fakers/overseas from saturating this channel as well?
reply
eclipxe 22 hours ago
Working on this actually
reply
NetOpWibby 18 hours ago
Where can we sign up for updates?
reply
ihsw 18 hours ago
[dead]
reply
Jotalea 9 hours ago
I really like the idea, as well as the "terminal" style the site has. however, I consider that an additional daily spend of $2 could be avoided. perhaps by caching common questions (like "what is this?"), or by using free tiers on API providers.

or, maybe I'm just too cost-conscious.

either way, the API limit is currently your "Achilles' heel", as it has already caused the bot to stop responding.

reply
sbinnee 22 hours ago
Nice. I had some fun. Good work!

One question. Sonnet for tool use? I am just guessing here that you may have a lot of MCPs to call and for that Sonnet is more reliable. How many MCPs are you running and what kinds?

reply
xeyownt 9 hours ago
> Automatic updates: Unattended security upgrades enabled.

Always wondered if such unattended upgrades are not security risk in itself, eg. seeing latest litellm compromise.

reply
dmazin 9 hours ago
Well, it should only update what it says: security updates (from official Ubuntu sources) unless you change the configuration.
reply
shreyssh 14 hours ago
Cool approach using IRC as transport. I've been experimenting with MCP as the control plane for letting AI agents manage infrastructure specifically database operations. The lightweight transport idea is underrated vs heavy REST APIs.
reply
ForHackernews 12 hours ago
This reads like it was written by AI. I don't understand how it provides any real security if the "guardrails" against prompt injection are just a system prompt telling the dumber model "don't do this"
reply
mobilefriendly 11 hours ago
I had the same thought as well. The firewall is just assuming a dumb model can't be tricked
reply
consumer451 20 hours ago
The demo seems to be in a messed up state at the moment. Maybe it's just getting hammered and too far behind?
reply
johnisgood 20 hours ago
Yeah, should probably implement rate-limiting. HNers were wildin'. :D
reply
consumer451 20 hours ago
Working better now. But, what just happened with that inappropriate link from nully?

Is handle impersonation possible here, or was it worse than that? Or, just a joke?

reply
oceliker 20 hours ago
Someone snatched the username when the actual nully left.
reply
Henchman21 20 hours ago
IRC without nickserv, good times
reply
consumer451 20 hours ago
That's pretty darn funny. The impostor should have given some believable responses to keep it going.
reply
johnisgood 20 hours ago
It was hilarious.
reply
greesil 18 hours ago
How do you keep it from getting prompt injected?

Oh I get it the runtimes are nice and small, you're using Claude for the intelligence. Obv

I think I'm just impressed with anthropic more than anything. Defcon would have me believe that prompt injections are trivial

reply
jaboostin 19 hours ago
lol I sent this link to my Claude bot connected to my Discord server and it started converting with nully and another bot named clawdia. moltbook all over again. I’m surprised how effortlessly it connected to IRC and started talking.
reply
agnishom 20 hours ago
> The model can't tell you anything the resume doesn't already say.

Good observation. But I would worry that in the scenario when this setup is the most successful, you have built a public facing bot that allows people to dox you.

reply
anoojb 19 hours ago
I wonder if this brings back demand for IRC clients on mobile devices? ;-)
reply
messh 19 hours ago
Can be significantly cheaper on a vm that wakes up only when yhe agebt works, see for e.g. https://shellbox.dev
reply
ruptwelve 18 hours ago
While I am a huge fan of IRC, wouldn't be simpler to simulate IRC, since you are embedding it? Or is the chatroom the actual point? Kudos on the project!
reply
mememememememo 21 hours ago
Yeah that chat got hosed by HN as any Show HN $communicationchannel does
reply
ekianjo 19 hours ago
But relying on a Claude API so you don't really "own the stack" as claimed in the article...
reply
selcuka 18 hours ago
Aren't LLMs commodity products these days? It's the same thing as running this on a $7 VPS that you don't "own".

I don't think switching to a different provider, or running an open one locally would affect the response quality that much.

reply
ekianjo 18 hours ago
The LLM is the key element here, not the 7 dollars VPS... The model itself has cost billions of dollars to train and of the service shuts down or is interrupted for some reason your fancy setup breaks like nothing.
reply
selcuka 18 hours ago
> The model itself has cost billions of dollars to train

But that has nothing to do with this use case, right? By the same logic, Linux has millions of man-hours went into it but we can use it for free on a $7 VPS.

> service shuts down or is interrupted for some reason your fancy setup breaks like nothing

No, it doesn't. That's what I meant by commodity. You can switch to another service and it will work just fine (unless you meant that all LLM providers might cease to exist).

Also note that they have a $2/day API usage cap, meaning that they are willing to spend $60+/month for the LLM use. If everything else fails, they can use those funds to upgrade the VPS and run a local model on their own hardware. It won't be Sonnet-4.6-level, but it will do. It just doesn't make sense with current dollar-per-token prices.

reply
chatmasta 17 hours ago
> The LLM is the key element here

No, the key (novel) element here is the two-tiered approach to sandboxing and inter-agent communication. That’s why he spends most of the post talking about it and only a few sentences on which models he selected.

reply
iammrpayments 14 hours ago
That was very educational, I found out I didn't know a lot of stuff.
reply
appstorelottery 16 hours ago
Lol. /nick The IRC implementation needs to be a bit more locked down. EDIT: So much fun to be in an IRC chat room - replete with trolling! Like a Time Machine to the 90's!
reply
iLoveOncall 22 hours ago
The model used is a Claude model, not self-hosted, so I'm not sure why the infrastructure is at all relevant here, except as click bait?
reply
jazzyjackson 21 hours ago
It’s not that deep, show HN is just that, show and tell, I seriously doubt this was built just to get engagement on social media
reply
petcat 22 hours ago
Meh it's kind of interesting. Even if it is just a ridiculously over engineered agent orchestrator for a chat box and code search
reply
echelon 22 hours ago
We need more infra in the cloud instead of focusing on local RTX cards.

We need OpenRunPods to run thick open weights models.

Build in the cloud rather than bet on "at the edge" being a Renaissance.

reply
abhishekayu 10 hours ago
Interesting setup.

The IRC part is neat, but the tiered inference is what stood out.

How do you decide when to escalate from Haiku to Sonnet?

reply
AgentTax 3 hours ago
Love the sandboxing design. The A2A passthrough where ironclaw borrows nullclaw's inference pipeline is a neat trick — one API key, one billing relationship. Curious how they are logging the split between Haiku and Sonnet spend per session. Once agents start running at any scale that attribution gets messy fast.
reply
adshotco 4 hours ago
[dead]
reply
ozozozd 17 hours ago
Super cool! Love seeing IRC in the wild.

Kudos and best of luck!

reply
password4321 12 hours ago
This looks like a fun project. I'm going to be that guy and spam this reminder regarding the HN submission text:

Don't post generated/AI-edited comments. HN is for conversation between humans

https://news.ycombinator.com/item?id=47340079

At the very least prompt your LLM to skip the AI-isms for "your" comments!

reply
topaz0 18 hours ago
Curious, which API key are you using?
reply
callamdelaney 8 hours ago
What on earth is the point? This is like saying you’re running wordpress on a vps? So what?
reply
eric_khun 21 hours ago
that's so fun ! how do you know when to call haiku or sonnet?
reply
Imustaskforhelp 15 hours ago
I have a 7$/yr vps 512mb ram which can run this. I have run crush from the charmbracelet team on the vps and all of it just works and I get an AI agent which I can even use with Openrouter free api key to get some free agentic access for free or have it work with the free gemini key :-)
reply
jgrizou 22 hours ago
Works very well
reply
tc1989tc 18 hours ago
it's great project
reply
m00dy 20 hours ago
Did you give your email access to a AI provider ?
reply
heyitsaamir 20 hours ago
Great idea and great write up!
reply
slopinthebag 20 hours ago
I can tell it's vibe coded because it takes about 1 minute for a message to appear.
reply
consumer451 19 hours ago
He had to put rate limits on it as it was getting hammered to hard by HNers.
reply
pugchat 10 hours ago
[dead]
reply
reachsmith 9 hours ago
[dead]
reply
leontloveless 9 hours ago
[dead]
reply
sudeepsd__ 13 hours ago
[dead]
reply
maxbeech 10 hours ago
[dead]
reply
commers148 9 hours ago
[dead]
reply
amoswang92 7 hours ago
[dead]
reply
johnwhitman 14 hours ago
[dead]
reply
craxyfrog 21 hours ago
[dead]
reply
agentpiravi 21 hours ago
[dead]
reply
cestivan 6 hours ago
[dead]
reply
johnwhitman 19 hours ago
[dead]
reply
teamorouter 17 hours ago
[dead]
reply
teamorouter 17 hours ago
[dead]
reply
sayYayToLife 20 hours ago
[dead]
reply
eu_93 12 hours ago
[flagged]
reply
felixagentai 22 hours ago
[flagged]
reply
getverdict 18 hours ago
[flagged]
reply
hackeman300 18 hours ago
Are you an ai
reply