Lucena Coder @LucenaCoder

Tokenmaxxing is for suckers. Try @LucenaCoder, a free, token-efficiency obsessed agentic coding harness. No login req, just add a key! 💻 Mac, Win, Linux. LucenaCoder.com Joined June 2026

Tweets

499
Followers

92
Following

356
Likes

4K

I'm sold. 1 MCP server. Contains the skills right in it. Paste into any agent I want and they have the same toolkit. It's token-efficient, and I can dynamically change what shows as needed. It works with WebMCP but there are like a handful of sites at this point with that.

0 0 1 80 0

0 0 0 13 0

View Details

Lucena Coder @LucenaCoder

a day ago

@ShelbyVentures Underboss is absolutely amazing and we can't wait to integrate it deeper into @LucenaCoder Loving this partnership. Need to get a fedora to make it official.

0 0 1 18 0

View Details

Lucena Coder @LucenaCoder

a day ago

Inference is killing lazy, token-inefficient startups. Fixed that for you.

0 0 3 9 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@Alok619308 Ohh we didn't want it for JUST us, we were looking at if it belongs in a recommended catalog for design for our @LucenaCoder users and our friends at @ShelbyVentures

1 0 1 32 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@delali This is the way our advanced stuff with @PoweredByEvok works. It builds single responsibility groupings really well. Meanwhile half of vibe coded apps are just one giant App.tsx now. LLMs really fight it naturally.

1 0 2 14 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@shaikhmudassir_ @MiaAI_lab @Zai_org Yes. It's a harness problem.

0 0 0 4 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@yongfook Users pay their way at the best available routing rates for the model they want. Even our BTC packs it's just based straight on model consumption rates. Can't expect a miracle there. The savings comes from the token efficient architecture instead.

0 0 0 3 0

View Details

Lucena Coder @LucenaCoder

2 days ago

If you OSS and nobody shows up, how do you feel? This is what I see happen most when founders OS their work before building something with at least a few hundred or thousand users. Like just fast forward through the scenarios in your head and see how you feel. Then decide. You can always OS later too.

0 0 1 32 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@ZaiforStartups Need. More. TPS! 🔥🚀

0 0 1 211 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@CtrlAltDwayne This is cool!

0 0 1 14 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@rayandabbagh I believe it. Have you used them? They are abstraction monsters. Code for the sake of code. Greenfield solutions painted with workarounds on day 1. Who wants to pay a premium for that.

0 0 1 26 0

View Details

Lucena Coder @LucenaCoder

a week ago

And we're just getting started...

1 1 8 2K 2

View Details

Lucena Coder @LucenaCoder

2 days ago

@johnroodepic @rileybrown Looking for another domain for memory? We have a couple fs like UserMemory and UserCache I think. Both dot com.

0 0 0 2 0

View Details

Lucena Coder @LucenaCoder

2 days ago

@rileybrown We are using mcplocker.com internally right now but we don't do separate skills. They should add that or just add an MCP with skills you want exposed?

0 0 0 2K 0

View Details

Lucena Coder @LucenaCoder

4 days ago

Best design skills MCPs?

0 0 1 38 0

View Details

Lucena Coder @LucenaCoder

5 days ago

@zachperret I think what people really want are "on demand workflows" when they think of things. Not so much agents just finding shit to do.

0 0 4 305 2

View Details

Lucena Coder @LucenaCoder

5 days ago

@andywalters Love love love this.

0 0 2 126 0

View Details

Andy Walters @andywalters

5 days ago

Normal people do not want to build AI Agents. If I have learned one thing running an AI startup that’s sold and delivered millions in contracts, it’s this. No one cares. No one cares to learn a new mental model. They just want to talk to a thing and get results. They don’t care about agents, skills, connectors or any of the other neologisms we in tech fetishize. They just want the thing to work so they can do something else with their lives. The mistake I see so many companies—and competitors—making is rushing into an ill thought through human computer interaction paradigm where one human manages teams of purpose built agents. I hate that future and we are actively building the opposite at AgentPress. Our ideal is the agents and the neologisms and the gizmos fade into the background and our customers experience AI as a singular ongoing conversation with a chief of staff who also happens to be a mentor, a friend, and a genius. I have always believed the true promise of generative AI is that it radically elevates the quality of our relationships with machines, lowering friction and increasing understanding. I see very few SaaS companies who seem to understand any of this. It is not enough to put a chatbot in your app. If you are not converting your entire company to selling a conversation between your superhuman AI and your customer, you don’t get it. — If this resonates with you and you’re on a sales team of 3-10 people and you’re drowning in follow ups, emails, and CRM updates, we are taking on selected design partners for our AI Chief of Staff for Account Executives.

65 16 308 39K 273

View Details

Mark Ajzenstadt @mardehaym

5 days ago

A harness is everything between a raw AI model and useful work. The model is the jet engine, and the harness is the cockpit: everything between raw thrust and a safe landing. I fly planes. You can buy the same turbofan Airbus puts in an A320, but that won't fly 180 people to LA. The engine is a commodity, but the cockpit is how you harness thrust into a destination. The failure patterns and recovery logic we encode into harnesses for our clients won't show up in an open-source repo, because that's judgment earned from real deliveries, not documentation. The argument that smarter models make harnesses obsolete is a bad take. A more powerful engine needs better avionics, not fewer, and you'll hit orchestration problems only your harness can solve as labs ship smarter models. The harness is where your alpha compounds.

Sherwood @shcallaway

6 days ago

There is no alpha in building your own agent harness. The best techniques will get discovered, copied, and eventually incorporated into open-source harnesses like Pi. A few months after that happens, these techniques will be made obsolete entirely as labs release smarter