anthropic Claude Code Releases · · 2.1.169

Claude Code v2.1.169: Troubleshooting, Directory Navigation, and UI Fixes

sdkaiengineerfinance
patch feature

This update to Claude Code introduces a `--safe-mode` flag for troubleshooting and a `/cd` command for changing directories mid-session without disrupting the prompt cache. It also includes numerous bug fixes addressing issues with UI responsiveness, enterprise policies, and agent management across platforms. These changes benefit developers and architects troubleshooting complex setups or working with large codebases, offering a more stable and efficient experience.

  • Add `--safe-mode` flag for troubleshooting
  • Implement `/cd` command for directory changes
  • Add setting to disable bundled skills and hide them from the model
  • Fix Up/Down arrow behavior in command history
  • Enforce enterprise managed MCP policies correctly
Features (2)
  • Add `--safe-mode` flag for troubleshooting

    A new `--safe-mode` flag has been added, which can also be set via the `CLAUDE_CODE_SAFE_MODE` environment variable. This mode starts Claude Code with customizations like CLAUDE.md, plugins, skills, and hooks disabled, aiding in troubleshooting.

  • Implement `/cd` command for directory changes

    A `/cd` command is now available to move a session to a new working directory. This allows users to change directories mid-session without breaking the prompt cache.

Enhancements (1)
  • Add setting to disable bundled skills and hide them from the model

    A `disableBundledSkills` setting and `CLAUDE_CODE_DISABLE_BUNDLED_SKILLS` environment variable have been introduced. These allow hiding bundled skills, workflows, and built-in slash commands from the model's view.

Fixes (3)
  • Fix Up/Down arrow behavior in command history

    Up/Down arrow keys now correctly navigate through wrapped rows of long input lines before recalling command history. History recall now enters at the near edge of the visual row.

  • Enforce enterprise managed MCP policies correctly

    Fixed enterprise managed MCP policies (`allowedMcpServers`/`deniedMcpServers`) not being enforced in various scenarios, including reconnects, IDE-typed configs, and cold starts.

  • Address UI stall on macOS for claude.ai users

    Resolved a ~30-50ms UI stall observed at the start of each turn for macOS users authenticated with claude.ai credentials.

Read the original announcement →

https://github.com/anthropics/claude-code/releases/tag/v2.1.169