Daily AI news, sourcedUpdates / Agents & coding
Updates Agents & coding · Official · 225 items · newest issue first · page 1 of 12
AreaAllModelsAgents & codingApps & productsChips & infraResearchPricing & limitsBusiness & fundingPolicy & safetyOther
- Cognition makes Devin Oncall available to all organizations
- Notion rebuilds its AI agents with their own computer and day-long tasks, and moves to a new credit system
- Cua releases the Cua Spaces SDK to run Claude Code, Codex and other agents on computers you choose
- Lovart 2.0 adds Arti, a personal design agent with memory and scheduled design tasks
- Amazon Bedrock adds retrieval instructions to agentic retrieval on managed knowledge bases
- LangGraph CLI 0.4.37 makes serverless the default deploy target
- Qwen Code 0.25.3 lets hosted agents launch read-only child agents and adds SSH, macOS and Kubernetes workspace runtimesOfficialQwenLMOct 19Agents & coding
- Epic Games' Unreal Engine skills plugin lets Claude Code control Unreal Editor over MCP
- OpenAI's desktop app now lets you edit previous messages within a thread, Tibo says
- Google Cloud AI Research open-sources Harness², a framework that improves an agent's harness for unseen tasks without changing model weightsOfficialGitHub google-research/harness2Oct 19Agents & coding
- Cline Desktop can now run tasks in the cloud, each in its own sandbox, while your laptop is closed
- Replit introduces Bring Your Own Cloud, letting enterprises run Replit and host published apps in their own Azure, Google Cloud or AWS tenant
- Cognition's Devin adds multiple side chats per message, editable queued messages and GitLab /devin commands
- DeepSeek opens its in-house search engine to developers with new Search and Fetch APIs
- Perplexity open-sources SPACE, the Firecracker microVM sandbox that runs every Perplexity Computer task
- Cline CLI 3.0.73 fixes GPT-6 requests on OpenAI and Azure endpoints and restores Claude Sonnet 5.5 as the default
- Replit adds Claude Haiku 5.5 to the Power Mode model selector for Core and Pro builders
- Cursor adds Meta's Muse Wonder, which scores 46.1% on CursorBench at default effortOfficialCursorOct 17Agents & coding
- Google DeepMind says four Gemini 4 Argon agents beat Minecraft from scratch in 27 hoursOfficialGoogle DeepMindOct 17Agents & coding
- Claude Code team explains how it reviews agent-written code, with merged PRs doubling and reverts down over 40%OfficialAnthropicOct 17Agents & coding