Cloudflare made Python Workers generally available on Sept 21, so FastAPI, Django, and Flask can talk to R2, D1, Workers AI, Queues, and the Python MCP package without a JavaScript conversion layer. Native C/C++/Rust packages still need PyEmscripten wheels.
Released Nov 4, gemini models and list models list the models your credentials can reach as text or JSON (ids, display names, token limits) without starting the TUI, so scripts can pipe the output cleanly. The same build also bridges the IDE companion into container sandboxes over exec stdio so diffs and open file context work inside the sandbox…
Admins add shared connectors and users add their own, so the coordination agent and the coding agents it runs can read tickets, CI status and internal docs; only remote Streamable HTTP or SSE servers work, not local stdio ones, and destructive tools start disabled (docs). The same Nov 2 update lets you keep chatting while workspace runs continue in the background…
Subagent status line updates, refreshed every 30 seconds, are now written by Haiku 5.5 from short notes instead of re reading the whole conversation, costing about 99% less each; Anthropic says sessions with background subagents use about 5% less of the limit. It's live in the CLI from v2.1.316 for most Pro, Max and Ultra users and all Team and Enterprise accounts (Claude Code changelog)…
Built on Thinking Machines' Inkling Small (276B total, 12B active), it's available now in Devin Desktop and the Devin CLI, free for Pro, Max and Teams plans through Nov 30. There's no public API, per token price or weights yet; Cognition says it's building a program to share weights with partners and researchers…
OpenAI plans to fold Pierre's Git backed Code Storage service into Codex; Pierre's open source projects stay open and existing Code Storage customers keep access after closing. Terms weren't disclosed and the deal still needs regulatory approval (Axios)…
Spotted something we missed? Start a thread.