@claudeupdates.bsky.social📦 Claude Code v2.1.283
Admins can now block models with `deniedModels` and pin exact versions via `availableModelsMatch`; auto mode becomes default on third-party providers, plus `/doctor prompt-audit`.
Full changelog + AI insights → claudeupdates.dev/version/2.1.283
@w512.bsky.socialGPT-6 Sol vs Claude Opus 5.5 🥊
Same 3 tasks, each model in its native tool: Codex vs Claude Code.
An evolution sim, a real code review, and more.
One came out ahead, but it was closer than I expected.
@cogtie.bsky.social
@nicolaounicos.bsky.social
@cdevmkt.bsky.socialstumbled on microsoft/playwright-mcp — MCP browser for coding agents, uses the a11y tree so no vision model needed to click. anyone using it with Cursor?
@programmerhumor-io.bsky.social
@shirtsfrenzy.bsky.socialWho Is JSON — Bold type asks WHO IS JSON on a deadpan developer t-shirt for programmers, vibe coders, and AI-tool users. Ships to US & Canada.
@rosgluk.bsky.socialLLM Systems: Operational Reference for Self-Hosted Inference
#AICoding #LLM #AI #Dev #OpenSource #SelfHosted #llama.cpp #GGUF
@donporque.bsky.social¿Por qué la IA está dejando sin ilusión laboral a los programadores? #InteligenciaArtificial #IA #Programadores #Programacion #DesarrolloSoftware #Tecnologia #Trabajo #EmpleoTech #Codigo #AICoding #Desarrolladores #felizviernes #25deseptiembre
@siva26.bsky.social👉 Visualpath’s training is designed to give learners
✔️ Date : 26/09/2026 @8:00PM IST
👉 Trainer Name : Mr. Arpan ghosh
🔗 Join Link: bit.ly/4havdNo
✔️ Meeting ID: 469 721 425368636
✔️ Passcode: Zy7gu2T7
📞 For Details: +91 7032290546
@hishnash.bsky.socialWe've had many questions about how SwiftFairy, our new code review app for Swift and SwiftUI, compares to skills and fits into an agent workflow. I've written a post explaining how agents work with it and why it's more reliable:
nilcoalescing.com/blog/Introdu...
#SwiftUI #AICoding #iOSDev
@juliangoldie.bsky.socialClaude Opus 5.5 just made a MASSIVE coding leap. ⚡
Agentic coding reportedly jumped from 52% to 66%.
Now it can build apps, animations, playable games, and complex projects from a single brief.
And it uses fewer resources while doing it.
@iamleeg.bsky.socialBuild 5 of Axminster just hit TestFlight! Lots of bug fixes and a few new features, including a whole new diagram type! chironcodex.com/axminster
#axminster #aicoding #softwareengineering #modeldriven
@claudeupdates.bsky.social📦 Claude Code v2.1.282
Big stability pass: resumed sessions keep extended thinking, 400 errors from undecryptable web search or redacted_thinking blocks are fixed, and managed settings get tighter.
Full changelog + AI insights → claudeupdates.dev/version/2.1.282
@agentpalisade.bsky.socialFuXi is a Go terminal coding agent that treats the LLM as a replaceable part, routes across providers with cost awareness and automatic failover, and describes its benchmark as a small, self-run task set: https://github.com/fuxicodex/Fuxi
#AiCoding #Terminal #golang
@courionai.bsky.socialGrok 4.7 ships at the same 2 and 6 dollars per million tokens as its predecessor. 71 percent on DeepSWE, and an oddly high score on a legal benchmark.
courionai.com/news/2026-09-21-grok-47-longer-reasoning-same-price
@eicker.bsky.socialThe widespread use of #AIcoding tools is causing a #crisis among #programmers, who are becoming overly reliant on these tools and losing their ability to write code by hand. This #overreliance is leading to a decrease in #criticalthinking and #problemsolving skills, resulting in a rise in #errors…
@nitor1122.bsky.socialSame model, different orchestration = 21% spread on SWE-bench.
5 teams independently built the same fix for AI coding drift.
Coincidence, or a clue about what really drives reliable AI-assisted development?
Read the blog: zurl.co/GrgtQ
@courionai.bsky.socialAnthropic's new Claude Code limits are a permanent 25 percent increase over the spring baseline, and 17 percent below what users actually had last week.
@courionai.bsky.socialAnthropic published what agentic coding did to its own engineering: eight times more code per quarter, ten times more tests, 25 times the testing load in six months.
courionai.com/news/2026-09-16-anthropic-ci-load-25x-agentic-coding
