Angehefteter Tweet
Matt Lee(Secret Third Thing)
2.2K posts
Matt Lee(Secret Third Thing)
@mattleefc
Building @tortoise_on_fc. mattlee on Farcaster
NYC Beigetreten Şubat 2019
1K Folgt804 Follower
Today, we're launching the @CoinbaseDev CLI and MCP server.
These tools give AI agents instant access to Coinbase's APIs for payments, trading, wallets, onchain data, and more.
Coding agents like Claude and Codex can now trivially integrate crypto's most trusted APIs. (1/3)
English
So excited to build stuff and play music at Kismet Casa and Farcon in a few weeks
The kismet is strong
Kismet Casa@KismetCasa
Kismet Casa is coming to Rome! April 26 — May 10 We’re bringing 11 artists and developers to live and create together, leading into a series of activations during Farcon All made possible thanks to @base Design by @tinyrainboot
English
English
@garrytan We think this is an overactive abuse-detection system, digging in. Also working on making the terms for -p crystal clear.
English
This seems messed up actually - when do the boundaries stop moving?
Anthropic only allows subscriptions with a real human pressing enter? You're going to have to verify with FaceID?
Peter Steinberger 🦞@steipete
Anthropic now blocks first-party harness use too 👀 claude -p --append-system-prompt 'A personal assistant running inside OpenClaw.' 'is clawd here?' → 400 Third-party apps now draw from your extra usage, not your plan limits. So yeah: bring your own coin 🪙🦞
English
@Austen High opportunity cost to not just spend my time building cool stuff
English
If you are potentially interested in applying to attend Gauntlet AI and have not, why not?
English
Guy who spends all his time researching and engineering complex abstract systems:
“Why would the general public be upset that they are increasingly forced to interact with complex abstract systems giving people like me more and more power over them?”
English
@bcherny @GergelyOrosz Let us use harnesses man we want to create freely
English
@GergelyOrosz Hmm there were no behavioral/model changes recently. If you see something weird (eg. the model refusing to do something) mind running /bug and posting the feedback id? Feel free to tag me
English
Anthropic really is burning more and more dev goodwill
Claude Code is suddenly getting unusable for stuff you could use it before (as in a day before!) and the AI now refuses to so stuff that it doesn’t think is strictly to do with software development.
No transparency why ofc
Theo - t3.gg@theo
Claude Code is basically unusable at this point. I give up.
English
“If ai can make perfect music in seconds how will human musicians compete?”
You just answered your own question bro
English
@praxisnation Why not use @xmtp_ for messaging?
English
HIRING — Praxis is an online community starting a new city. We are looking for:
1. ARCHITECTURAL DESIGNER — to concept residential and public spaces in the city.
2. APPLIED CRYPTOGRAPHER (messaging) — to securely connect our global community.
3. AI ENGINEER (multi-agent systems) — to connect our global community through new social experiences.
To apply, send your resume, GitHub/portfolio, and a short note on your interest in Praxis to HIRING @ PRAXISNATION dot ORG.
English
@0XunoYou @farcaster_xyz We’re all having fun on Farcaster
English
Posting hot takes about nyc on twitter is the most desperate attention seeking behavior a modern human is capable of
English
@lapislagoons I’m not like the other blue guys tho
English
"Tech job openings rebounded sharply in 2026, challenging popular narrative that AI is wiping out engineering roles...more than 67,000 software eng job openings, highest level in 3 years. Listings have doubled since a trough in mid-2023." businessinsider.com/ai-isnt-killin…
English
the lobsters are at the gates and anthropic doesn't know they can climb
English
@Romy_Holland relfection selfie, artsy photo, psychosis
English
My mom’s century plant is blooming and she just sent me the most random hauntingly beautiful message
English
@brexton I don't recommend giving it access to that but also yeehaw
English
Notes from day 3 of trying to claw-pill myself as a historically loud openclaw hater:
1. I think a lot of the addiction comes from how much it makes life feel like an RPG, you see someone post something on X like a new skill or Karpathy’s new wiki thing and you then have FOMO about “leveling up your agent”. Like tamagotchi or Pokémon in a way
2. I feel myself just wanting to point my agent/claw at endless things/repos that I want to add, but not verifying if it’s slop or not. Makes me nervous about the infinite software house of cards we’re building of pointing our slop against to use more slop code and producing even more slop code
3. I find a lot of the workflows I’m discovering “cool” but I don’t know if I’m convinced that it materially makes me more productive/efficient, or if it just gives me the *feeling* of being more productive/efficient. I’m now increasingly convincing myself that I need to use more tokens for things even though it costs me more money. That’s Jevon’s paradox I guess
4. My optionality paralysis has not gone away, in fact, I’m really annoyed by how many alternatives and tools there are now and if I need to assess those too
5. Claws are not the right form factor for enterprise. Too time-intensive to set up, and not multiplayer
6. I am nervous about all of these tools I’m giving it access to
English
@dexhorthy I am using claude-cli with my openclaw now it works
English
Concerning re: anthropic. The previous narrative just went out the window
Reports of openclaw usage with the plain sdk (as was supposedly permitted) now being blocked based on system prompt, even if using the claude agent sdk
I was previously a little to the anthropic side of the spectrum on this because of EXACTLY one argument - “third party harnesses don’t use caching properly, can’t be controlled with feature flags, etc”
If they are blocking use of the claude agent sdk wholesale in openclaw, then this completely invalidates that argument and I desire an answer as to what is allowed and why. I am disappointed that the communications thus far have failed to articulate the reasons here, and does make it harder to trust whatever they say next.
However I will maintain cautious optimism that there is a good explanation for all this beyond the cheap “rug pull” “evil” “kill all the startups” jeers
dex@dexhorthy
like I’ve said a few times, well within TOS to do this, they built the model, if they wanna give you inference at pennies on the dollar on the condition that you use their harness, great, they have the right to do this. On this topic in particular, I don’t understand the “evil” or “rugpull”, jeers. There was never any promise to give people cheap inference. Before the claude code max plan we were all paying per token to use this stuff. And we’re more or less happy to do it (sure the VC funding helps). Every enterprise I know pays per token because when you use subsidized inference, YOU are the product. “Have some cheap code, in exchange for helping to train the next gen of models” You can hate on that particular behavior if you want but nobody is making you take part in that particular market dynamic. Do I wanna see a world where model companies take some of their massive financial gains and use that to pull everybody up? Of course. I hope it happens some day. An allegory perhaps: If public e-bike company gave you a subscription on rides and you proceeded to around ripping out batteries and sticking them in your own bike and ride around town, you’d get banned for that too. Especially if your bike was poorly wired and overloaded the batteries/cause them to flame up etc. Banning that behavior would deliver far better results for the people who were using the system as designed
English