https://theagenticweekly.com/news/stack-diagram-agent-skill-architecture-diagramsAgentic Weeklyen2026-11-04T20:16:27Zstack-diagram, an agent skill that turns one sentence into an architecture diagram with official logoshttps://theagenticweekly.com/news/uncut-0-1-lossless-headless-coding-agent-harnessAgentic Weeklyen2026-11-04T20:16:21ZUNCUT 0.1 is a 7 MB headless coding-agent harness that never trims context or outputhttps://theagenticweekly.com/news/claude-issued-auth-plugin-auto-connect-betaAgentic Weeklyen2026-11-04T18:04:26ZClaude can now sign employees in to plugins itself: Claude-issued auth beta for Team and Enterprisehttps://theagenticweekly.com/news/llama-cpp-v0-9-0-fp8-mistral-large-4Agentic Weeklyen2026-11-04T14:02:42Zllama.cpp v0.9.0 adds FP8 E4M3 quantization, Mistral Large 4 and Chatterbox TTShttps://theagenticweekly.com/news/sarvam-prism-open-weights-decision-modelAgentic Weeklyen2026-11-04T13:05:08ZSarvam Prism: an open-weights decision model on Sarvam 105B that returns probabilities, not texthttps://theagenticweekly.com/news/tencent-hy4-vision-computer-useAgentic Weeklyen2026-11-04T11:04:48ZTencent releases Hy4 with vision and computer use, at $0.834 per million input tokenshttps://theagenticweekly.com/news/dify-1-19-0-nested-workflow-tracingAgentic Weeklyen2026-11-04T10:05:42ZDify 1.19.0 adds nested workflow tracing, portable Workflow Tools and Skill commands in difyctlhttps://theagenticweekly.com/news/agent-autoloop-1-0-background-cli-agentsAgentic Weeklyen2026-11-04T03:04:43ZAgent Autoloop 1.0 runs Claude Code and Codex in the background, asking only when you must decidehttps://theagenticweekly.com/news/comfyui-v0-44-0-minimax-h3-prores-partner-nodesAgentic Weeklyen2026-11-04T01:17:57ZComfyUI v0.44.0 adds MiniMax-H3 support, ProRes video saving, a VAE dtype node and new partner nodeshttps://theagenticweekly.com/news/vercel-security-review-public-betaAgentic Weeklyen2026-11-03T22:06:35ZVercel Security Review: Vercel Agent now scans every GitHub push for vulnerabilities (public beta)https://theagenticweekly.com/news/fireworks-forge-ultra-fast-nexus-desktopAgentic Weeklyen2026-11-03T21:07:48ZFireworks Forge: Ultra-fast inference preview, half-price Background Mode and Nexus Desktophttps://theagenticweekly.com/news/anthropic-message-threads-claude-codeAgentic Weeklyen2026-11-03T16:59:43ZAnthropic's message threads beta cuts Claude Code time to first token about 35% on long sessionshttps://theagenticweekly.com/news/github-copilot-jetbrains-airAgentic Weeklyen2026-11-03T16:11:52ZGitHub Copilot is now an agent in JetBrains Air, with cloud tasks billed to your Copilot planhttps://theagenticweekly.com/news/cloudflare-agents-0-33-0Agentic Weeklyen2026-11-03T15:05:26ZCloudflare Agents SDK 0.33.0 moves the Pi harness to Pi 1.5, cutting storage writes, with no rollbackhttps://theagenticweekly.com/news/hume-audio-decision-apiAgentic Weeklyen2026-11-03T13:07:55ZHume's Audio Decision API answers plain-language questions about a recording with probabilities, no transcripthttps://theagenticweekly.com/news/open-webui-v0-13-0Agentic Weeklyen2026-11-03T13:01:34ZOpen WebUI v0.13.0 adds a tabbed workspace with terminal, browser, desktop and Git tabshttps://theagenticweekly.com/news/anthropic-provable-inferenceAgentic Weeklyen2026-11-03T07:00:27ZAnthropic open-sources FlashProve, a zero-knowledge prototype that proves which model wrote a responsehttps://theagenticweekly.com/news/phone-harness-mcp-cloud-iphones-androidsAgentic Weeklyen2026-11-03T06:02:56ZPhone Harness MCP server lets Claude, Cursor and Codex agents drive cloud iPhones and Android emulatorshttps://theagenticweekly.com/news/letter-letterpress-1-growth-engineering-coding-agentAgentic Weeklyen2026-11-03T02:10:24ZLetterpress-1: Letter's beta coding agent that opens growth and SEO pull requests on git-hosted siteshttps://theagenticweekly.com/news/gemini-2-5-pro-flash-deprecated-auto-route-december-1Agentic Weeklyen2026-11-03T02:10:10ZGemini 2.5 Pro and 2.5 Flash deprecated: API calls auto-route to newer models from December 1, 2026https://theagenticweekly.com/news/deep-principle-hydrogen-1-0-open-weight-science-modelAgentic Weeklyen2026-11-02T21:05:00ZHydrogen-1.0: Deep Principle's 753B open-weight science model, post-trained from GLM-5.3 under MIT