fucory

1/ Smithers is the best Orchestrator and you should try it What is Smithers Smithers is an Orchestration framework where you declaratively create a DAG of Durable Tasks as xml and reactively regenerate parts of that DAG as Tasks complete How do you use it You use it as a CLI or a agent skill. Your agent then writes, updates, and monitors those scripts What special features does it have - MDX for prompts - Library of prebuilt 0-config workflows - All workflows represented as code that is easy to modify, customize, or compose into larger workflows - Built in observability with SQLLite and pormetheus - Supports Claude code, Codex, Gemini, and Kimi subscriptions - Supports all harnesses including Claude Code, Codex, Pi, and Opencode - Supports Agent SDK Why is it so good? The main reasons Smithers is so good is 1. Agents are great at writing complex workflows because Agent's are great at React 2. It is thoughtfully built with @EffectTS_ Durable execution primitives enabling more reliability for very long running agentic tasks 3. Powerful CLI for watching and interacting with wrokflows 4. MDX is a really great existing standard for expressing prompts 5. The declarative nature of Smithers and it's great observability makes it very easy to read and quickly reason about What is best way to get started? bunx smithers init in an existing project and try it out Alternatively check out the docs smithers.sh

@jaivin This is Solid.js in Rust not React though

react in rust, is this cursed

Antigravity is now tabslop

@eniwhere_ @pvncher Yea @pvncher is absolutely right. @mattpocockuk is shaping his way of working to the tool. Don't do this. Use smithers.sh and shape the tool to your way of working

@pvncher @FUCORY he needs the Smithers pill

If this is true, then we need something other than skills to handle single track multi step workflows, that are not composable and can’t be auto invoked by models.

Smithers is a harness sdk. You use it to easily build a custom fitted harness for a problem. Excited to ship benchmarks this week

@irl_danB Why draw on elements of react when you could use literal react Smithers.sh

later this week I'll be excited to release a new kind of harness we've been building it's one suited particularly well to the OpenProse paradigms of: * declaring intent in structured language * pushing all further decision making into the agent I found that with almost all of the OpenProse systems that we'd been writing, the missing piece was something like an event-based architecture to add dependencies across runs, one that looked something like a data-flow graph between OpenProse Responsibilities we built a prototype of this on in our private cloud before deciding to pull the guts of it out into a relatively simple sdk, which we'll be open sourcing we've started calling it a Reactor harness, because it draws on elements of React to be efficient at keeping a composed world-model up to date with respect to events and upstream changes this harness is arguably an RLM, or if not a fully generalized RLM, draws on many RLM-class principles in its architecture, whereby many layers of ReAct sessions are themselves being managed by meta-sessions, where context state is passed--as in the original OpenProse release--via pointer rather than in-context, and where every instantiation of an agent (reactor session or fulfillment session) is in an environment where it has access to a shell that includes the full system state as a git repository that it can write code to interact with in the same way that React manages dependencies between declarative components, re-rendering only those that need update, the Reactor Harness manages dependencies between "OpenProse Responsibilities" (declared ideal world model state) in keeping with with the current OpenProse tenets, the re-render is still a question that we push into the intelligent agents, so it's more of an "intelligent memoization". in further keeping with our tenets, all intent authoring/specification still happens at the OpenProse Framework layer: declared markdown contracts and (optional, imperative) ProseScript to dictate agent behavior inside of any given session we're taking ideas for benchmarks for long-running systems like this, though we're also designing some of our own in absence of others, stay tuned for results initially I was going to wait until we have benchmarks to release this, but I think it's worth just getting a rough version out to collect feedback before declaring some hill-climbed numbers that don't pass the community vibe check, so please vibe check and send feedback bookmark this tweet and grok will kindly put the release tweet in your feed later this week. and if you want to try an early-waste-of-time release candidate, send me a DM (h/t @raw_works and @josemontesdeoca for major contributions here)

@EricForgy @PubKey I’m here with flashbots team and bunch of other teams come by sometime early next week

Hello my parasocial friends 👋 I'll be in NYC all this week (finally) learning about consensus and hanging around @PubKey as much as possible. I'm happy to meetup, but you might get an earful about Multiswap 🤠 DMs open.

Humans need a better memory solution

@SecDataResearch If your thought was correct it would be in the pretraining and you could just have claude elaborate

@FUCORY False. I cannot elaborate due to ongoing engineering.

If an agent skill becomes popular enough you won't need the skill anymore because it will end up in the pretraining

Had some skill issues publishing it's on npm now

Smithers 0.21.0 is here Antigravity CLI for Google agents. Gateway operator console for runs and approvals. Injectable sandbox providers with diff bundles. Workflow eval suites. Stdin JSON input. Nested sandbox boundaries. Process-backed sandbox runner. Hosted control-plane. Keyboard-driven demo deck. smithers.sh/changelogs/0.2…

@mattpocockuk @bcherny Hitl works better with a higher level substrate that supports multiple harnesses like Smithers.sh

@bcherny It also screws up all HITL checkpoints in skills, making it largely unusable

People often ask what my biggest tip is for getting the most out of Claude Code. These days my #1 tip is: use auto mode Auto mode means no more permission prompts. It is the key building block for multi-clauding: start a session, then while it runs, work on another session in parallel.

@FUCORY Swear on my life I’m not another tabslop

Here is what every app looks like now This app is called TabSlop TabSlop is like Crabs. Just as many animals evolve into a crab form, all apps are evolving into TabSlop TabSlop is not the future. Nobody wants to manage tabs. It scales to at most 10 to 20 tabs and agents should be managing that anyways What is your favorite TabSlop app?

@lajarre Point your llm at Smithers.sh/llms-full.txt tell it you don’t see the value prop and ask it how it could help you

@FUCORY stuck at lvl 3 for now, gimme the nudge

Why I'm so bullish smithers - People who actually try Smithers have a high chance of becoming power users What we need to improve about the funnel - More entrypoints than just me tweeting - Convincing people to give smithers a deeper look when they go wtf (show them value prop earlier But overall the bottlenecks being top of funnel and Smithers performing so well once people give it a try is a good problem to have

@merlinaudio_ @Chinese_XU @croloris Really strong points

@Chinese_XU @FUCORY @croloris Banning ski masks because criminals wear them often does nothing to combat crime, nor will criminals care about the ban. But going skiing now sucks for everyone

The Zig compiler project rejects AI contributions for very concrete reasons. - The contributor pipeline is a core aspect of our business model and produces value FAR BEYOND the code. Breaking it is an existential threat to Zig. kristoff.it/blog/contribut… - For some projects it's a problem to rely on vibecoded infrastructure. Some have learned this the hard way with Bun recently. Zig is general purpose programming language for building critical infrastructure and we want to give maximum freedom to our users. See quoted message.

@theo Man that sucks unbelievable and unfair how much of this crap one is subjected to if they are a public figure

If you’re curious why more people aren’t speaking up about the green card news, it’s probably this.

@benswerd Absolutely! Smithers.sh will be bringing you a lot of users soon so I promise it will pay off

@FUCORY You haven’t seen anything yet. also I like your takes. Want 10k credits?

Resist the call of the cloud. Build on bare metal.

@benswerd Fwiw as a datapoint I built a fully working version of what you are offering with firecracker and I'm still planning on using freestyle because the price difference still makes sense to have someone focusing on that piece of the product more than I have bandwidth for

@FUCORY You make a good point. Resist other clouds, build on bare freestyle.sh/products/vms

@davis7 You should try out smithers.sh I think it's cursed enough to get you to the next psychosis level Imagine BAML but with orchestration. And that orchestration is React

It's a meme but I really have been letting the psychosis take over as much as possible to figure out what I can actually do with these things - codex desktop app computer use spam - hermes agent that I've been loving - going as hard as possible on parallelizing in projects - making as many of my tools into markdown files as humanly possible - cloud agents + "cloud" agents that are t3 code instances on my mac mini - letting codex entirely control and setup my computer exactly how I want it to - unironically using gbrain, it's quite nice lol - seeing how far I can push local models on my 5090 (qwen 3.6 is very good) doing the real network setup was expensive, difficult, but was a great decicion. I feel so much more comfortable going harder with this stuff having A) a real firewall and B) hard gates around the mac mini and NAS so in the worst case scenario they can't compromise anything else on the network (the firewalla was an incredible purchase) and I still don't think I'm going hard enough