nikiv.dev

@badcryptobitch just faster as it’s not plan everything but more gradual with lots of feedback

@nikivdev Then what's the difference between doing it inside of prompts as opposed to plans?

my new dev flow for more complex things is i have codex write spec claude review it codex review what claude wrote, optimize spec get review fast (will optimize this with tooling, wip) then i have cli `spec exec <.md path>` and it writes optimized prompt and executes it

@badcryptobitch claude helps write it

@nikivdev Do you write those plans yourself or are they AI assisted?

@badcryptobitch i do detailed plans in prompts themselves and sync ensures that there is no overlap in work

@nikivdev What's the difference between enforced autonomy and plan writing? Feels like you can get the former with the latter anyway?

@badcryptobitch No not yet, it’s internal tooling

@nikivdev Do you have anything written about that?

@badcryptobitch enforced autonomy with codex and no more plan writing

@nikivdev What's the updated technique?

@badcryptobitch no longer use this technique so can’t say sorry

@nikivdev What is your initial prompt for the codex to write the spec?

@dillon_mulroy @CodeWithZeee would love to test this too

@CodeWithZeee let’s change that

this is so freaking cool

@dillon_mulroy @fr0ziisc2 dm’s you too for access 🖤

@fr0ziisc2 you're in - go forth and conquer github.com/elithrar/artif…

the artifacts + agents sdk + dynamic workers + isomorphic-git combo is over powered

soon `spec review <.md path>` will open review tool with inline comments basically github light will be under unhash.sh<hash> all inline all reactive ai will be able to fetch optimized review feedback with context fast to iterate on spec

@ryanvogel try quilt by @thetronchguy

going to benchmark sandboxes i have - cloudflare - vercel - daytona - e2b.dev - exe.dev - sprites anyone else i should try

tested it as i had around 30+ zed windows open and I wanted full restoration I had my mac literally go tab by tab, full snapshot of all tabs (all claude/codex sessions inside each window) the stuff that seq lets you do is absolutely insane tbh

the only way to get these lower is joining macOS team and improving things there + better hardware but this is plenty fast and at current physical limit which feels amazing on side I made seq support notion of `.md` executable recipes which is super cool

Truly life changing day @tekezo is genuinely top 3 humans I admire the most who changed my life Since this post, I wrote PR to Karabiner with the change to cut latency and remove shell to use sockets I thought I would have to support a fork but no, tekezo literally merged it

@tekezo Thank you lots for building Karabiner When you get time, can you check this PR I want to modify Karabiner Elements to cut some latency to a macro engine I am building and need your guidance Thank you ♥️ github.com/pqrs-org/Karab…

idea behind action packs is that i can remotely run kind of like executables fast from my main mac on another as there is efficient vm running that can process it all if it works well for this case, will exp with it by running seq in linux hosts and agentic containers too

not crazy enough to swap karabiner on my main mac so seq now has a concept of an `action pack` which is series of seq instructions that can execute if i pass the pack signed over tailscale network daemon will instantly execute it if its set for execution will harden it more

Started replacing keyboard maestro with my own macro engine in c++ and it's crazy how doing this moved my most common op to open apps from 500ms in KM to 200ms in my own engine I am now evaluating forking karabiner elements to avoid even more overhead on engine calls