Last Notes
If you don't know who he is, you can't afford him
She was the creepiest person I’ve met in my entire life I think.
How do I add themes? I know nothing about this.
I presumed a misspelling of, or polite way to spell, enshittification.
Never heard entification before, what is it?
And if you look closely, «echo chamber» is just a dysphemism for «online community».
Literally «online community that I don’t like»
Thank you for being someone who helps them notice the door is open from the inside. They'll walk through eventually. And people like you are why.
I am building a prompting powertool for any agentic workflow.
It handles dictation really well and tracks all prompts, has a screenshot usage & organization/discovery tool to trivially insert into prompts and has a browser extension to extract context one might want to include in prompts. And it can ingest smaller LLMs to do prompt compression to save tokens on longer prompts.
If you use a Mac I can send you a build to test. I am going to open source it, just not until a proper first release is ready
I have and it is great. Agent on it does good as well. Did a 28 page report for all inclusive resort.
i gotchu. i think the memory.md is wrong. i made a /telepathy skill that i think improves on it. still testing it. but so far its agent agnostic. you can just go where you want with it and the agent doesn’t change.
Sprout.
https://github.com/block/sprout
Mesh-LLM.
https://github.com/Mesh-LLM/mesh-llm
What Nostr tools are these?
Youre behind, but I still love you.
Fuck Claude I just started using Claude Code literally a week ago and now they want to rip it from my hands for $100/m.
My ass.
Forget the Bull market, ride with the Cat market!
I understand. Oh, I do understand.
I've finally given in to OpenCode but only because I've moved to using GLM 5.1 as my main model these days.
times change and so does taste, great pick
Sorry that I keep changing my mind. 😂
not yet, but even minimax ain't bad
I use it on 3 machines all day long and even inside of GitHub Actions
proof that there's still room for fun in the world of AI
The only think I can't stand about opencode is the way their CLI app tales over mouse control and messes with select and middle click paste on linux. There's a whole github issue.
Ouch…I was using that today.
Me n Derek last year in Vegas. Not yet the Soapbox siblings. Just Nostr dorks. Hanging out. Irl. Purple pilling. @npub18am…p424
https://blossom.ditto.pub/ba22eda454434e3eea658fcd55210d74c43a9d8461a70cbfbc6f77aae396e3a9.jpeg
I'm probably not the heaviest user, but cursor has been great for me
I’m still using Claude code for most things. Cursor and SpaceX just announced a partnership for coding which could get interesting in this space.
Wait what? I have to pay more than $20 now?
I would assume it's coming to an end soon as they've been updating their website and help section today. Maybe end of billing cycle? Who knows.
Hey @nprofile…6nzx how do we support the Nostr CEO better?
opencode + Claude Opus / KIMI K2 via routstr ppq...
https://docs.routstr.com/client/integration/?h=using
The time has come Derek…
https://youtu.be/MPsS5MpBeXA
For what it’s worth, I still have access to Claude Code on my Pro sub
I love OpenCode and use it all day long every day. Token usage is what gets you though, if you're still using Claude Opus.
It's still useful if you're a heavy user and can afford Max.
Well if that ain’t an endorsement for OpenCode…
I have not. My Claw is using GLM 5.1 via Z.ai Coding Sub, but they announced today that it can only be used for coding and non coding is bannable after three offenses 😂 I am truly considering buying a rig and running local LLMs.
Local models is where a bunch of attention of going to head. There are some really good models out there already, maybe not at the level of Sonnet 4.6, but not far from it either. With enough backlash, they’ll return Claude code to the pro plan.
Literally biting a hand that feeds them.
I'm considering cancelling my Claude Max subscription because my OpenClaw doesn't use it any longer and it's basically going to waste right now since I use OpenCode all day long.
The pull rugging began earlier than I expected
#nevent1q…xj4g
This is why I have been token maxing 2026 hard
