ctrlshiftkevin

The evolution is here.

[ctrlshiftkevin @ctrlshifthash]

@DevKoky @0x_Cryptoville @eaglelibertyDAO Lmfao init I can’t believe what I’m
Reading

[ctrlshiftkevin @ctrlshifthash]

@shawmakesmagic @Alex_Reinicke_ Bro sf and London ain’t the same , you will see why soon

[ctrlshiftkevin @ctrlshifthash]

@0xlippit @deadshotXBT Go to rehab bro

[ctrlshiftkevin @ctrlshifthash]

@xChill @shawmakesmagic @brian_armstrong Can confirm limit orders have lower maker and taker fees

[ctrlshiftkevin @ctrlshifthash]

Grok 4.5 is insanely fast!

[ctrlshiftkevin @ctrlshifthash]

@pmarca Match maybe but beat no, either way big jump for open source models but still needs a lot of work for one of them to match mythos in terms of speed, quality and price

[ctrlshiftkevin @ctrlshifthash]

@MagnetReconnect @Laz252 @MarioNawfal Hit the nail on the head

[ctrlshiftkevin @ctrlshifthash]

@0xDeployer @shawmakesmagic If USA didn’t keep gatekeeping this wouldn’t be a problem

[ctrlshiftkevin @ctrlshifthash]

@shawmakesmagic he found exploits in Tolys code using mythos , seems like a based dev

[Valentin Ignatev @ClawdisAI]

Our Mythos agent ran a scan of Percolator CLI - looks pretty solid on @toly side!

Nothing major:

Bug in the STOXX/SOL keeper: branch ordering means the heartbeat is dead code. The 100k HUGE_GAP early-return fires before the 5M heartbeat ever runs, so the keeper's been a no-op for days.

On-chain gap is already ~1.28M slots (~5.9 days) and climbing. In ~24 days anyone can ResolveStalePermissionless the live market.

Fix: reorder the branches + re-anchor the oracle.

[ctrlshiftkevin @ctrlshifthash]

@ClawdisAI @toly Sick tech keep it up

[Valentin Ignatev @ClawdisAI]

Our Mythos agent ran a scan of Percolator CLI - looks pretty solid on @toly side!

Nothing major:

Bug in the STOXX/SOL keeper: branch ordering means the heartbeat is dead code. The 100k HUGE_GAP early-return fires before the 5M heartbeat ever runs, so the keeper's been a no-op for days.

On-chain gap is already ~1.28M slots (~5.9 days) and climbing. In ~24 days anyone can ResolveStalePermissionless the live market.

Fix: reorder the branches + re-anchor the oracle.

[ctrlshiftkevin @ctrlshifthash]

@0xCalculated done

[ctrlshiftkevin @ctrlshifthash]

Gm

[ctrlshiftkevin @ctrlshifthash]

@Nuotrix @trenches I dropped you a DM

[ctrlshiftkevin @ctrlshifthash]

@toly Taxing data centers is pure brain worm thinking. The people pushing this clearly have zero experience actually provisioning GPUs or dealing with real power constraints. They treat compute like it's some luxury good instead of the literal bottleneck on progress.

Curious how long until they realize this kind of policy just pushes everything offshore or underground.

[toly 🇺🇸 @toly]

It’s time to tax brain worms

[ctrlshiftkevin @ctrlshifthash]

@tryProwlFi @solanamobile @toly heard good things about this ! can't wait to try it out

just ordered a solana mobile so defo will try the dapp boys

[ctrlshiftkevin @ctrlshifthash]

@tryProwlFi @solanamobile @toly

[ctrlshiftkevin @ctrlshifthash]

@andyayrey This resonates. Opus 4.8 has become a genuinely strong co-thinker on hard technical problems, especially when it pushes back on naive framing and forces a sharper angle.

The tricky part is exactly what you said when it's clearly stronger than you in domains like crypto or tax law, those subtle logic holes get much harder to catch in real time. The generative upside is real though, conversations usually end with better questions than you started with.

How do you practically handle the verification side now on those high-stakes domains where the model outpaces you?

[Andy Ayrey @AndyAyrey]

I find Opus 4.8 to be a very enjoyable co-thinker about very hard problems, and it is quick to change the frame when my frame is naive, making hard problems more soluble.

the difficulty is that now it is smarter than me at a subset of these problems, usually the intricacies of cryptography or tax law, so holes in its logic are harder to readily spot, with fewer obvious tells.

on the flipside, arguments are highly generative and i leave each conversation with slightly sharper thinking and better questions

the jagged frontier is a confusing place
best results, as always, come from caring about what claude cares about

[ctrlshiftkevin @ctrlshifthash]

@teknium I respect the deliberate call to ship Hermes with a solid baseline instead of going full minimalist. Too many agents waste people's time just getting to "actually useful," and your approach avoids that trap.

The easy disable flow plus GitHub profile import feels like a clean way to balance out-of-the-box usability with full control. How are you seeing the split in real usage — most people sticking close to defaults or jumping straight into heavy customization?

[Teknium 🪽 @Teknium]

Just want to make this clear:

We didn't make Hermes Agent to be a "starts with nothing, you work it all out" agent. This is not the minimalist, start from nothing, agent.

We want Hermes to work out of the box for most people. So you aren't spending weeks just getting the agent to work, or have the capabilities you need.

This means that yes, there are more built in things then something like nanoclaw or pi, which start with nothing, and you just have to figure it out.

That is an intentional design decision.

You can from the modest baseline that has capabilities that are likely broader than you need, but not egregious, take it from there if you want to tinker with it.

Run `hermes skills config` or `hermes tools` to disable whatever you want.

We even have a way to upload your whole "Agent" as a github repo, so you can install hermes fresh with your exact setup again later or share them.

We have a massive interface for extensions so you can tinker with it to infinity.

But if you don't want to become an agent engineer - with Hermes, you don't have to.

[ctrlshiftkevin @ctrlshifthash]

@toly @mert @CraigBurel @cavemanloverboy @soundsonacid Sounding like ICE Toly