favurdev

You need to define an escape hatch for the LLM otherwise it will loop on increasingly small issues and even start making up issues that don't actually exist. A simple "fix all medium and high severity issues" guidance will prevent low-priority nits from consuming all your tokens.

Its an interesting idea, typically the rules are defined while bootstrapping and carry from one project to the next since personalization of the rules is hard and tedious. I could see larger orgs using this to bootstrap a fully customized lint rule file. But its a shame its only typescript!

Actually like the minimalist design - congrats on the release. Appears pretty useful especially while claude is so opaque with sub agents

When we create a large amount of memories or research docs I always find it best to create a table of contents or a reference doc citing all the docs and a short sentence about it.

Then that TOC becomes the "required read" before everything in any new agent. Which increases the memory doc read substantially since it understands what each doc is natively.

Congrats on shipping this. Going all-Swift can't have been the path of least resistance.

Curious what made you take it - was it mostly about how the thing feels to use day to day, or something you kept hitting with the Electron ones?

We took loop engineering a step further and run the agents inside workflows to keep them on task across the whole 2 day run. Happy to get into how it works.

Took me longer than I'll admit to figure out what I had to hit!