tontinton

Everyone copies https//maki.sh. I've been the first!

In this day and age there's no excuse to not build a new distributed system without DST, the AI is not lazy.

Can you also compare in the charts https://maki.sh?

Should give you some competition.

Oh very nice, can you also compare it to https://maki.sh?

Would be interesting to compare to a harness optimizing for cost reduction too.

Competition is good

Comes across as this "I'm special", and kinda cringe...

Or https://maki.sh :)

I was really confused when reading the article that their claim is you don't know formal verification, I was expecting something way juicer.

Is it really? I'll need to check it out

They can do both, the model decides when to use code_execution like any other tool

I've made https://maki.sh for use cases such as this

Are you freaking kidding me with YC throwing money at something like this? I guess I can fund raise just by having built https://maki.sh, and months ahead of other founders too...

All the harness tools have been re-implemented as plugins in Lua: https://github.com/tontinton/maki/tree/main/plugins

Here is the full API reference: https://maki.sh/docs/lua-api/

Mine sends even less - https://maki.sh

Yep this is a glimpse into the future of 500+ t/s, which is in my opinion the next big thing that validates Jevon's paradox (the models are already smart enough)

I meant harnesses / TUIs, sorry to confuse.

Using fd/rg sounds interesting, honestly it would require little tweaks to the bash tool lua plugin, either add to the description to prefer these binaries instead or something like that.

In general though I much prefer "advising" and encouraging the LLM to use the native tools like grep l/glob, they are implemented to be super fast, and you will get better parser output.

Create an issue! :)

It's a TUI you're right, but it's also a harness.

As much as I hate to admit, T the tools you provide, the descriptions, and prompts, all amount to pretty big changes in experience, even using the same models.

The biggest ones are: using tree-sitter to index code files as a tool, code_execution tool running a workflow of tools inside a python interpreter (monty), and not being a harness developed by the company profiting from selling you the shovels (and introducing "dynamic workflows" aka spawning 50 agents).

Enjoy, I can't go back to other agents now, too spoiled by the speed

Totally wrong, you underestimate the frontier's incompetence in anything other than building LLM models (ehm ehm flickering TUI for a year "written like a game engine").

I ran a bunch of benchmarks and there are proven ways to reduce tokens while achieving the same results (finding the same CVEs / finding the same bugs in CRs, etc...).

See https://maki.sh, it's my own little proof.

Yep exactly my thoughts, went and looked at the code for the deepseek provider in my coding agent. and basically all of what the author wrote there is implemented... http://github.com/tontinton/maki for the curios

What's your moat?

Yo that's really similar to my very own https://github.com/tontinton/maki only I'm MIT and you're GPL, cool

Also easier to ship a binary like a cli

or you can use a non vibe designed efficient Rust TUI coding agent made by yours truly, all my coworkers use it too :) called https://maki.sh!

lua plugins WIP