On November 20 (UTC), Anthropic shipped Claude Code 2.1.330. It adds sign-in to organizations managed in the Anthropic Enterprise Admin Console, applying the organization settings and its cloud provider or gateway. A new built-in easel mod draws charts on request and under MCP metrics results. Recurring scheduled tasks and /loop now run up to 30 days instead of 7, and more rm commands aimed at the root, home, a top-level folder or the working folder now ask for approval, or are refused when nobody can answer. Self-hosted runners gain --start-dir and two latency metrics, and the VS Code extension adds /fork and /workflows.

Key Takeaways

  • ✓Enterprise: after claude auth login, sign in to organizations managed in the Enterprise Admin Console; Claude Code applies the organization settings and uses its cloud provider or gateway
  • ✓easel: a built-in chart mod that draws charts on request and under MCP metrics results, on Pro, Max and, where available, Team and Enterprise on macOS and Linux; turn it off in /config
  • ✓Scheduling: recurring scheduled tasks and /loop now run up to 30 days instead of 7; tasks made before this release still end at 7 days
  • ✓Safety: more rm commands aimed at the root, home, a top-level folder or the working folder now ask for approval, and are refused where nobody can answer
  • ✓Self-hosted runners get --start-dir and /metrics latencies from work order to registration and first session; fixes include multi-minute freezes after compacting long chats, WebFetch hangs and ExitWorktree deleting another session's worktree
🧭

Heavy Claude Code use: compare subscription limits and API bills

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

🔬

In-Depth Technical Analysis

Background

Claude Code is moving from a personal terminal tool to something organizations manage centrally: they want one sign-in, model routing and permission policy, and long-running agent tasks need steadier scheduling and stricter guardrails. Claude Code 2.1.330, released November 20 (UTC), addresses those areas.

How it works

After claude auth login, users can sign in to organizations managed in the Anthropic Enterprise Admin Console; Claude Code applies the organization settings and uses its configured cloud provider or gateway. A new built-in easel mod draws charts when asked and under MCP metrics results (Pro, Max and, where available, Team and Enterprise, on macOS and Linux; it can be turned off in /config). Recurring scheduled tasks and /loop now run up to 30 days instead of 7, and more rm commands aimed at the root, home, a top-level folder or the working folder ask for approval, or are refused when nobody can answer.

Numbers that matter

Anthropic published no benchmark figures. This is a tooling release focused on enterprise access, scheduling and stability: it fixes multi-minute freezes after compacting very long conversations, WebFetch hangs and ExitWorktree deleting a worktree another session was using, and it reverts 2.1.329's removal of the automatic retry when max_tokens exceeds a limit.

What it means for developers

Admins can have members sign in through Admin Console organizations for a single configuration, and teams running unattended jobs should note that risky rm commands are refused when nobody approves them. Self-hosted runners can start in a repository subfolder with --start-dir and track dispatch latency with the new /metrics series, and VS Code users can try /fork and /workflows.

Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

Up NextAnthropic Launches Claude Meetings Beta: Claude Joins Google Meet Calls as a Visible Participant, Builds Docs and Whiteboards Live and Writes the Notes, Rolling Out to Team and Enterprise
Read Now→