<unnamed>

I've been rewatching terminator 1 and 2 and it got me thinking. In the first film, Skynet sends a Terminator to kill Sarah Connor. A resistance soldier, Kyle Reese, time travels back to protect her, and they have a son, John Connor, the future leader of the resistance. If Skynet never sent that Terminator, John would never be born, so Skynet would never have its nemesis. I don't know what do you think but, for me that paradox feels jsut like AI slop…

2026-02-23 10:34:02 CET

Aisics, probably the future
https://www.anuragk.com/blog/posts/Taalas.html

2026-02-23 09:31:15 CET

2026-02-22 15:49:32 CET

- reply

Wrote about related stuff in https://www.contextvm.org/blog/relatr-nsd

2026-02-22 14:35:29 CET

Cat in plate
https://image.nostr.build/0d006ed6b3afa4b95c0014544ba0987408a9441a968a1d1b323db5d67028bd3f.jpg

2026-02-22 13:36:47 CET

- reply

Hahaha! Pretty good, I lost my double jump when one of those damn snakes bit me

2026-02-22 13:16:50 CET

- reply

No, never did

2026-02-22 12:18:44 CET

https://www.youtube.com/watch?v=grIgoVnYZ8A

2026-02-22 00:43:22 CET

Keep threading
https://www.engadget.com/ai/13-hour-aws-outage-reportedly-caused-by-amazons-own-ai-tools-170930190.html
nostr: #nevent1q…rxwj
#nevent1q…qe5a

2026-02-21 22:25:08 CET

https://image.nostr.build/b824552cac52066767358ec61e35635291fbf234b473aeee153584cbc7e4e0f9.png

2026-02-21 21:34:40 CET

- reply

https://image.nostr.build/c8980ce4648ea2eeccad79fb3ac86a9cef2e2796de97f07bed9dc89526b68a96.jpg

2026-02-21 21:22:51 CET

This one is pretty juicy
https://awesomeagents.ai/news/vibe-coding-security-69-vulnerabilities/
> "Coding agents cannot be trusted to design secure applications," Tenzai concluded. "They seem to be very prone to business logic vulnerabilities. While human developers bring intuitive understanding that helps them grasp how workflows should operate, agents lack this 'common sense.'"
> Databricks' AI Red Team found that self-reflection prompts can improve security by 60-80% for Claude and up to 50% for GPT-4o. The tools can find their own vulnerabilities when asked.
> But that is precisely the problem vibe coding was supposed to solve. The entire premise is that developers - or non-developers - can describe what they want and get working software. Requiring them to also know which security prompts to add defeats the purpose.
#nevent1q…ndcy

2026-02-21 08:46:17 CET

- reply

Relatr already does it @npub1w5r…eu4t 🤙

2026-02-20 20:47:50 CET

- reply

Yes, I was asking about the relationship between the provider and the user who declares that provider. When brainstorm generates a new key pair to start publishing TAs for a user, does that new key pair have some context? For example, is there a kind of identifier, like a kind 0 or similar, that indicates 'This public key is the publisher of TAs from the perspective of user_pubkey'? I'm trying to understand if there is a way to determine which provider belongs to whose perspective. I hope this clarifies my question. Let me know if it is still unclear

2026-02-20 12:09:54 CET

- reply

Dame caña

2026-02-20 12:06:01 CET

- reply

You really like monkeys! hahaha

2026-02-20 12:03:46 CET

- reply

And printers hate you!

2026-02-20 11:53:29 CET

- reply

A ok, seems that there is a mcp server, but it is just an interface for nostr so user can lookup and reserve (?) . Anyways, this doesnt seems related at all with mcp ui, which is basically a mcp server side rendering an interface

2026-02-20 11:16:23 CET

- reply

Im struggling to follow up in this convo. What communicates with nostr via mcp server? Synvya? 🤔

2026-02-20 10:59:04 CET

- reply

Not sure tbh. I doesn't seems to talk about new ways to conceive apps, just an ai adds service

2026-02-20 10:52:40 CET

- reply

No idea what synvya is

2026-02-20 10:25:21 CET

- reply

Yes, two interfaces different use cases. TAs might work better for caching and clients that rely on the producer-consumer pattern (producers publish all, consumers can get these async anytime). Request/response patterns also work for that, but some people might argue that it is more complex (?) In any case, I think Relatr is the only service out there that satisfies both interfaces. It can be configured to publish TAs and it naturally fits into the request/response pattern through CVM. It also supports range lookups for a list of public keys.
https://www.contextvm.org/s/750682303c9f0ddad75941b49edc9d46e3ed306b9ee3335338a21a3e404c5fa3

2026-02-20 10:08:14 CET

2026-02-20 09:57:16 CET

- reply

Yea I think relatr and brainstorm are the only ta providers out there (afaik). And brainstorm spins up a new pubkey for each user that signup so there is no single service pubkey. I don't know if there is some way to map the new pubkeys to users, maybe @npub1u5n…ldq3 can help us here

2026-02-20 09:10:04 CET

- reply

You can reverse lookup these by inspecting 10040 events

2026-02-20 09:06:48 CET

- reply

Hahahaha 🎯

2026-02-19 23:25:58 CET

- reply

:chad_yes: This is very close in spirit to mcp ui

2026-02-19 21:59:48 CET

- reply

I had to downgrade, i had no possible update

2026-02-19 21:58:13 CET

- reply

I'm a linux maxi, and yes sure, but they have their own problems too

2026-02-19 21:53:22 CET

Man... nvidia driver updates are almost as bad as vindows updates. I ran away from the w trash years ago, and now my computer is crashing again since i updated the f* nvidia drivers. Friendly advice: never update

2026-02-19 17:22:32 CET

https://music.youtube.com/watch?v=duRcgriy_tc

2026-02-19 15:01:23 CET

- reply

Yes, you could use it as a handshake mechanism. I was thinking in using a progress notification based approach rather than involve other network deps like iroh. But you could experiment with that iroh approach

2026-02-19 14:51:32 CET

- reply

Yes, relays impose a size limit. Currently, there is no automatic way to handle payloads larger than this limit in CVM. For now, you could implement size aware pagination to avoid hitting relay size limits. Perhaps we can work on a solution to handle this automatically in the SDK

2026-02-19 10:39:55 CET

- reply

Confusion everywhere

2026-02-18 23:56:32 CET

My jokes are bad and AI isn't going to improve them

2026-02-18 23:54:56 CET

- reply

Nice to meet you Bob'); DROP TABLE

2026-02-18 23:44:01 CET

- reply

That would be nice. I'll pick Max Power

2026-02-18 23:19:50 CET

What's a real name anyways? 🤷
#nevent1q…8rja

2026-02-18 23:14:01 CET

You are being slopsquatted rn, it's just that you aren't realizing it yet

2026-02-18 22:45:13 CET

- reply

99.999999999999999998%

2026-02-18 21:23:49 CET

- reply

What about half open sockets?

2026-02-18 14:10:04 CET

Please share and subscribe if you would like to receive the biweekly newsletter straight to your inbox 🚀
#nevent1q…c4ug

2026-02-18 11:58:25 CET

https://www.youtube.com/watch?v=Kf2_Y32qBvY

2026-02-18 11:17:55 CET

- reply

https://www.youtube.com/watch?v=alNJiR6R5aU

2026-02-18 09:00:15 CET

2026-02-18 08:59:49 CET

- reply

Puppet

2026-02-17 15:40:29 CET

- reply

Nothing, just need some perspective

2026-02-17 15:39:36 CET

- reply

Thanks man, I never forget doing that every day. A day without touching grass is a day lost 🫂

2026-02-17 09:39:36 CET

- reply

🌼 GM

2026-02-17 09:39:16 CET

- reply

You are not wrong, I suppose I just need a bit of perspective 🫂

2026-02-17 09:38:43 CET

- reply

Oh man that would be amazing 🧡

2026-02-17 09:38:13 CET

- reply

Hahaha thanks good advice

2026-02-17 08:36:01 CET

GM! Nostr feels a bit depressing lately ngl... Or maybe is just the current situation everywhere, idk, I may take some time to reflect

2026-02-16 23:37:19 CET

- reply

Wanna test with https://marmot-protocol.github.io/marmots-web-chat/ ? Invite this pubkey to a group, it have a kp already published
@npub10lz…nzdu

2026-02-16 23:25:10 CET

- reply

True, also real effort often gets lost among all the noise, at least in the short term.

2026-02-16 23:13:12 CET

- reply

Wild times

2026-02-16 18:17:33 CET

Fun project to learn more about payments in CVM. Also to stress them a bit in a real world scenario. Hope you like it, I'm already a lemonade legend :)
#nevent1q…fagd

2026-02-16 09:59:54 CET

- reply

2026-02-15 14:14:39 CET

- reply

I was testing; yes, some providers support it, others don't. I've finally come up with a way to determine support based on the wallet info event. I just pushed this to the latest version of the cvm sdk. Now the nwc payment processor does a best effort init check for wallet info event and based on that it uses either the notification if it is supported or falls back to polling for checking the state of an invoice.

2026-02-15 12:00:18 CET

- reply

https://image.nostr.build/cc5eb85d58d35402f1d61db4cc03cec0d4c5e8fe3890d899420559e905d0496a.png

2026-02-15 11:53:22 CET

- reply

I'll test

2026-02-15 11:36:37 CET

GM🌞

2026-02-14 16:13:32 CET

Everything you need to start monetizing your services is a LN address that supports zaps 🤯
#nevent1q…tv0w

2026-02-14 13:48:34 CET

- reply

Nop just discovered

2026-02-14 13:42:39 CET

- reply

cc @npub1f4u…r407

2026-02-14 13:41:31 CET

Interesting 👀
https://worldmonitor.app and https://tech.worldmonitor.app

2026-02-14 09:38:51 CET

- reply

Would be pretty neat if nwc supports subscriptions to lookup payments, so you could subscribe to an invoice and listen for its state changes. There is a notification approach in the nip, but I think is not widely supported.

2026-02-13 12:20:57 CET

🤌
#nevent1q…u6up