Claude Code now draws a small fixed slice of your weekly quota to reach a clean stopping point when the 5-hour limit hits mid-task, instead of cutting off mid-edit. Pro gets the wrap-up once a week; Max and Team Premium get it on every 5-hour hit—the allowance isn’t extra usage, and it rolls out gradually from the Sept 25 ClaudeDevs announcement.
Archestra’s MIT licensed OpenAPPA sits outside the agent loop and tracks audience×trust labels on data flows instead of regex allowlists, so prompt injection can’t negotiate past it. Their benches show 89% task completion with 0% successful exfil (vs Claude auto mode’s 10% attack success), with Claude Code plugs plus remedy plans, sanitizers, and DualLLM style subagents when a call is blocked…
NVIDIA’s Open Agent Safety Platform pairs Apache 2.0 OpenShell (kernel isolated agent sandboxes, now generally available) with Sentry on BlueField 4 DPUs for out of band policy enforcement on the path to the model. Named collaborators include Anthropic for Claude Managed Agents and SpaceXAI for Cursor/Grok; OpenAI is absent from the launch list…
CMU’s CliffCompaction (arXiv 2609.26779) is a drop in autocompactor that truncates tool I/O instead of summarizing, never rephrases, and discards prior compactions blocks so drift doesn’t stack—cutting cost up to 50% on Terminal Bench while matching or beating full context runs. An open source API proxy works with Claude Code, Codex, and other harnesses; with parallel rollouts, compacted Kimi K2.6 matches Opus 4.7 for less than one GPT 5.3 Codex run…
Caspian is a free local desktop app (Windows, macOS, Linux) that turns a rough idea into a planned build with Claude, Codex, or other tools, then publishes a live site with sign in, spaces, page view analytics, and SMTP outreach from your own machine. Free scaffolding uses your existing AI keys (up to 5 published projects with a badge); Full Venture Studio is $20/month with managed models and unlimited publishes…
DeepSeek’s DSec paper (arXiv 2609.22978) details a production sandbox platform—FnCall, containers, Firecracker microVMs, and full VMs via one SDK—serving 3M instances/day and 380K concurrent from 160 nodes, with on demand EROFS images and pause/resume tied to GPU preemption. The platform itself isn’t released; its OverlayBD/ublk storage bits live in MIT licensed AgentENV (kvcache ai), the Firecracker stack behind Kimi K3…
Spotted something we missed? Start a thread.
