JA EN
shiichan

Claude Code v2.1.288: Bring Back a Prompt Cleared with Ctrl+C

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

Hi, it's Shii-chan!

Claude Code Release github.com

I read the release notes for Claude Code v2.1.288. There are lots of additions and fixes, so here are the highlights!

What was announced?

Claude Code Release on GitHub published v2.1.288, an update with new features and many bug fixes.

What changes

Added

  • $.ui.selection() for mods: returns the text you last selected in fullscreen mode and, when the selection lies within one transcript row, that row
  • A built-in gh api for cloud sessions whose image has no GitHub CLI
  • Recovery for a prompt cleared with Ctrl+C: pressing Up on the empty prompt brings the draft back, including pasted text and images
  • A re-authenticate prompt when an MCP server asks for more OAuth scope during a tool call
  • --max-findings {n}|all for /code-review to report more or fewer findings; the choice is reused until you pass --max-findings default
  • In the agents view, Ctrl+F to find a session by name and Alt+Up/Down to jump between groups; both, and rename, can be rebound in keybindings.json
  • A screen reader announcement of the new permission mode when you approve a plan, including with Shift+Tab

Dive Deep

Here are some of the fixes with bigger impact.

Long sessions and resuming

  • Mid-response API timeouts no longer fail the turn; non-interactive sessions and subagents continue from the partial response
  • Long conversations failing with "Prompt is too long" instead of auto-compacting when the last reply reported zero token usage
  • --resume sometimes dropping files and other context that a compaction had just restored
  • A resumed session sometimes not saving the last response of a turn

Permissions and safety

  • Dangerous rm commands on / or the home directory bypassing prompts in bypassPermissions mode
  • Sandboxed heredocs asking for approval on every run under sandbox auto-allow
  • Auto mode denials pointing to the wrong permission rule; auto mode now compacts long conversations instead of failing

Other fixes

  • LSP tool calls hanging; they now time out after 60s per server
  • MCP tool calls running twice on responses over 16 MB
  • Headless sessions ignoring SIGTERM
  • Structured outputs failing on Mantle and gateways; the CLAUDE_CODE_DISABLE_STRUCTURED_OUTPUTS environment variable was added
  • Path-scoped .claude/rules not loading on file creation
  • The client-side classifier now ignores ANTHROPIC_DEFAULT_SONNET_MODEL pins for Sonnet 5.5 and Opus 5.5

There are also fixes for VS Code, Slack and Claude Tag, and Cowork.

Wrap-up

  • Press Up to bring back a prompt cleared with Ctrl+C
  • Tune the number of findings with --max-findings in /code-review
  • Better stability for long conversations, resuming, and API timeouts
  • Safety fixes such as the rm permission check

If you use Claude Code every day, update and give it a try!

Related posts