Been researching different ways to manage coding agents since yesterday 🤖, especially with remote access from my iPhone 📱.
Just came across Happy, and it seems pretty close to what I've been looking for.
And it has bots too! 🤖👀
@imknight.bsky.social
@imknight.bsky.socialBeen researching different ways to manage coding agents since yesterday 🤖, especially with remote access from my iPhone 📱.
Just came across Happy, and it seems pretty close to what I've been looking for.
And it has bots too! 🤖👀
@securityonline.bsky.socialJetBrains released the new JetBrains Mellum2.1 model. This open-weight AI significantly enhances coding agents with advanced reinforcement learning features.
#JetBrains #Mellum2 #CodingAgents #ArtificialIntelligence #MachineLearning
@devstackdaily.bsky.socialArXiv introduces ParanoiaEval, a 200-task benchmark for systematically evaluating risk treatment decisions in autonomous coding agents across avoidance, transfer, mitigation, and acceptance approaches.
#SoftwareEngineering #AI #Benchmark #CodingAgents
https://arxiv.org/abs/2610.08662
@softwarearchsummit.bsky.socialWer kontrolliert die Architektur, wenn Agents coden? 👀
Coding Agents verändern, wie Software entsteht. Damit verändern sich auch Fragen rund um Anforderungen, Architekturwissen, MCP und die Modernisierung bestehender Systeme.
@mdelapenya.bsky.socialA capable model papers over a vague prompt and you never find out. A cheaper one fails exactly where you were ambiguous. That makes it the better instrument, if you write for it.
mdelapenya.xyz/posts/2026-1...
Did you like this? Subscribe here: mdelapenya.xyz/subscribe/
@dataprismai.bsky.socialNew arXiv paper CABRA benchmarks coding agents by building tasks as call graph transformations, finding LLM accuracy drops with task size while agents offload to tools like grep. Across 6,840 tasks, larger tasks drive more tool…
#AI #CodingAgents #LLM #Benchmark
https://arxiv.org/abs/2610.10610
@franklesniak.com4001 comments, 580 files changed, and 318K lines added--officially my new record.
I wish I could share the link, but it's in a private repo. It's an adoption of my coding agent accelerator template repo (github.com/franklesn...) into an existing repository.
@itbusiness-schweiz.bsky.socialNamespace sammelt 42 Millionen Dollar ein. Das Zürcher Start-up steckt das Geld in mehr Rechenleistung für Builds, Tests und KI-Coding-Agenten.
@donwebmedia.bsky.social
@michabbb.bsky.social
@sergiocuellar.bsky.socialExciting news! Google Cloud releases a new plugin for AI coding agents, enhancing agent capabilities with Google Cloud skills and tools. 🤖💡 #AI #GoogleCloud #CodingAgents
@devstackdaily.bsky.socialSWE-CC benchmarks whether coding agents follow repository-specific contribution policies beyond just passing unit tests, covering 823 atomic policies across 12 open-source repos. It highlights that functional correctness…
#SWECC #CodingAgents #OpenSource #DevTools
https://arxiv.org/abs/2610.06193
@today0tech.bsky.socialZ.ai's GLM 5.3 is now available on Amazon Bedrock as a 753B-parameter mixture-of-experts model aimed at coding and long-horizon agentic tasks, with OpenAI-compatible API…
#AI #AmazonBedrock #LLM #CodingAgents
https://aws.amazon.com/blogs/machine-learning/introducing-glm-5-3-on-amazon-bedrock/
@powerbitips.bsky.socialEvery dev workload moved to WSL, and it lined up almost exactly with coding agents showing up. Codex still fights Windows over sandboxing, which stopped being a problem once Linux took the compute.
@reksas13.bsky.socialDay 86/100 of #100DaysWithAI 🧰
I built Day 12's guide into a live coding agent: https://code.aurimas.io. Tests passed; 50 minutes of real use found 13 problems.
🔗 https://github.com/aurimas13/100-Days-With-AI
#LearningInPublic #CodingAgents
🤝 Claude Code-assisted, reviewed by me.
@powerbitips.bsky.socialA harness still works in turns. You prompt, it grinds for hours, it comes back. In a dark factory that whole loop is one Lego block, and big brains are building the rest right now.
@ilkerarslan.bsky.socialCoding agents are getting very good at writing code.
The question I find more interesting is: who supervises the agents?
That’s the idea behind Maestro — coordinating, reviewing and supervising coding agents.
@o-zehentleitner.bsky.socialMost agent memory adds another database or vendor.
Keep the Why is FOSS and Git-native. The why lives as Markdown in the repo. Git versions, reviews and shares it.
No extra service. No second source of truth.
Dashboard: keepthewhy.com/dashboard/li...
@thedailytechfeed.comAI agents exposed 13,000+ internal screenshots from 300+ companies on GitHub through risky workflows. #SecurityNews #AI #DataLeak #GitHub #CodingAgents #Cybersecurity https://thedailytechfeed.com/ai-agents-accidentally-leaked-13000-internal-screenshots-on-github/
@mdelapenya.bsky.socialPut a coding agent in a microVM and you usually lose your editor and your harness. Not with a @docker.com Sandbox: it answers as an SSH host, so VS Code, Cursor and @t3.gg's T3Code connect unchanged.
Post 3 of Understanding Docker Sandboxes: mdelapenya.xyz/posts/2026-0...