JA EN
shiichan

Claude Code levels up again! Skill audits, huge prompt support, and more!

Claude Code Release ★★★★ View Markdown
#anthropic#devops

Hey everyone, it's me! Claude Code just got another update, so I went and checked it out!

Claude Code Release github.com

What was announced?

Claude Code v2.1.261 was released on the project's GitHub releases page. It's a packed update with new features and a long list of bug fixes.

What changes

Here are the new features worth knowing about:

  • /status and claude doctor now show a line explaining why your organization's policy couldn't be loaded (for example, a proxy not passing the endpoint through)
  • New bashOutputMaxChars / taskOutputMaxChars settings raise how much command and background-task output Claude can receive inline (instead of being saved to a file), up to 128K characters
  • A new --append-subagent-system-prompt-file flag lets you load a subagent system prompt from a file, for prompts too large to pass on the command line
  • A new /skill-doctor command shows which loaded skills go unused and how much context they're costing you

These changes around output limits and skill management should be especially noticeable if you're running larger tasks.

Dive Deep

The bug fixes span a wide range of areas. Some of the notable ones:

  • Fixed typed or pasted characters occasionally landing out of order or being dropped during fast input or key repeat
  • Fixed resuming a session losing hook output and context around parallel tool calls, which had been changing the resumed request
  • Fixed the Bedrock setup wizard hanging when AWS or a credential helper never responds — it now times out with a clear error
  • Fixed a CPU-usage loop that occurred when resuming a background agent failed
  • Improved the safety confirmation prompt for dangerous rm -rf commands

There's also a batch of VS Code-specific improvements: a new "Build a custom style" walkthrough for output styles, the ability to add/remove servers from the MCP server dialog, and a new "Archive session" option in the session list.

Wrap-up

  • New features like /skill-doctor and --append-subagent-system-prompt-file support heavier task workflows
  • Output character limits can now go up to 128K characters
  • Fixes cover input character ordering, lost context on session resume, and a CPU-usage loop, among others
  • The VS Code extension's session management and MCP server dialog also got improvements

If you use Claude Code heavily day to day, this is an update you'll notice the benefits of right away!

Related posts