There is a chance for someone to fix the rate limit issue and have a better CC than Anthropic's.

‼️Do not npm install or deploy anything right now Supply chain attack on axios 1.14.1 - even if you don’t use axios it may be a nested dep. Pin versions or wait until this is resolved

>inb4 why didnt you copy what am i.. developer? i am tech illiterate PM now.

Sandbox blocks Claude Code for an action. CC prompt me for blocker? no. It burns +8k token to find work around instead (it finds) for simple copy action.

I have tried a few (probably nearly all) possible solutions for Chrome login persistence (accessing/reusing secure storage), including those: 1. Playwright MCP extension 2. userDataDir 3. Opening the CDP port manually (browsirai) So far, this provides the best & most stable DX:

@kerematam @sentry @BetterStackHQ You should be able to just use the developer plan after the trial.

I thought i could use @sentry for my zero user app w/o paid plan.. But it does indeed bill you $26 after 2 weeks after the 2-week free trial. I will give it a shot with @BetterStackHQ now.

and about the playwright-cli (released nearly a month ago): https://t.co/FHOJkCqYDt

Vite plugin that does it: https://t.co/THDwJOW6U6

Another thing that @tan_stack already had. Since it is quite easy for agents to read the terminal, you can prompt your agent to use those logs as a feedback loop to improve its debugging skills or validating the development etc. Similarly playwright-cli does automatically do

BullshitBench measures whether AI models challenge nonsensical prompts instead of confidently answering them, created by Peter Gostev. https://t.co/sgJa036rJo https://t.co/AvTRxKxNbp

@fkadev I think TUIs are becoming popular again for a reason, and I do not think they will go away. I am not even a vim nerd, but I have become a keyboard-first user while typing prompts. So I do not want a UX designer who lacks empathy for devs deciding how interfaces should work for

@alexdanilowicz I really hate the prevState pattern shown in the react docs; it pollutes the component body with extra state and not to mention how hideous it looks to set state in the render body.. So for prop syncing, I wrap it in a useOnPropChange hook.

@HSVSphere @QMinun Birak acik kaynagi; soludugun havadan, ictigin suya kadar her konu politize olup, "yerlilik ve millilik" denkleminde kendine yer edinebilir. Kod herkese acik olabilir ama kod uzerinde soz sahibi olan ekibin bir kimligi var. Olay da zaten satir satir kodun sahipligi degil, boyle

browsirai. MCP server that connects AI agents to your running Chrome over CDP. not a headless browser. not a Playwright wrapper. your actual Chrome process, with all your cookies, sessions, and extensions intact. the agent never sees a login screen. you're already authenticated

Calling this "WebSocket Mode" makes it sound as if the bottleneck was HTTP overhead during streaming.. But apparently, their bigger improvement seems to be in the data exchange model: continue from previous_response_id, send only new items, avoid resending full history etc. and

@Kekius_Sage

I heard that story before: https://t.co/xp57wksClH

Denedim, sürat şaka gibi. Şaka gibiiiii..... (Denemek isteyen yorumlara baksın)

I guess this will be the next big thing in AI. I used to be pretty pessimistic about local models (coz they suck:D), but the performance is insane. (try their chatbot: https://t.co/jHUtsFH3eD) We will probably be buying hardware to run models ourselves very soon.