Cloudflare opened a waitlist for fully managed Cloudflare OS agent workspaces (custom domain, Access policies, AI Gateway)—self-serve OSS deploy stays available. New agent capabilities include mounting GitHub repos to explore/fix/PR, broader Google Workspace Gatekeeper (Gmail read/draft/send, Drive), and exports to xlsx, CSV, PDF, Markdown, or HTML.
Microsoft and Hugging Face released ThinkingBox: 507 stateful business workflows graded on terminal backend state and side effects across 20 repeats—not tool call trails or final replies. Run it via OpenEnv; Claude Opus 5.5 leads pass@1 ( 67%), and 80% of failures are tool handling, not reasoning…
Cloudflare’s closed beta Monetization Gateway lets U.S. sellers charge AI agents per request for sites, APIs, MCP tools, or datasets using HTTP 402 / x402 (USDC on Base via Coinbase’s facilitator)—no checkout redirect. Early production paths include AI Gateway pay per inference, Ceramic.ai search, Stocktwits signals, and API2PDF…
Oh My Pi (OMP) 18.6.0 parses DeepSeek DSML tool call envelopes on every host—including llama.cpp, LM Studio, vLLM, and custom gateways—and strips unparsed DSML markup from history so a broken call retries instead of ending the turn. It also stops Antigravity Claude Opus 5.5 / Sonnet 5.5 from vanishing after omp models refresh (fallback client version now 2.19.1)…
DeepSeek’s MIT Harness (dsh) v0.2 public preview ships official Apple silicon Mac and Windows x64 desktop apps—Cordis “everything is a plugin” runtime with Creator mode (chat built plugins), file/diff sidebar, and scheduled Automation tasks. The installer bundles dsh (no separate Node/pnpm); web UI still via npx @deepseek ai/dsh web ; expect breaking changes…
Claude Code 2.1.289 (Oct 3) closes permission gaps: Read deny rules resolve symlinks, Bash deny/ask rules catch env prefixed and bare assignment compounds under sandbox auto allow, and managed deny/ask on nested shell parts hold over user mod approvals. Plugins get an agent.spawn hook plus idle/waiting states in $.agent.list()…
Spotted something we missed? Start a thread.
