Skip to content

v2.1.193

Choose a tag to compare

@marckrenn marckrenn released this 25 Jun 21:56
· 15 commits to main since this release

2.1.193

(☝ Click to see full diff.)

Highlights:

  • Bash and PowerShell commands now go through the auto-mode classifier to enforce uniform shell safety checks
  • Auto-mode records denial reasons in transcripts, denial toast, and /permissions for auditing blocked actions
  • Added live file-path autocomplete to bash mode (!) for faster, less error-prone file selection

CLI changelog

  • Added autoMode.classifyAllShell setting to route all Bash/PowerShell commands through the auto-mode classifier instead of only arbitrary-code-execution patterns
  • Added auto-mode denial reasons to the transcript, the denial toast, and /permissions recent denials
  • Added claude_code.assistant_response OpenTelemetry log event containing the model's response text. Redacted unless OTEL_LOG_ASSISTANT_RESPONSES=1; when that var is unset it follows OTEL_LOG_USER_PROMPTS, so deployments that already log prompt content will start receiving response content on upgrade — set OTEL_LOG_ASSISTANT_RESPONSES=0 to keep prompts-only.
  • Added live file path autocomplete to bash mode (!)
  • Added a startup notice when MCP servers need authentication, pointing at /mcp
  • Added automatic memory-pressure reaping for idle background shell commands (disable with CLAUDE_CODE_DISABLE_BG_SHELL_PRESSURE_REAP=1)
  • Fixed /model and other client-data-gated UI showing stale/empty state immediately after /login
  • Fixed backgrounding (←←) spuriously cancelling with "N background tasks would be abandoned" when all running tasks carry over to the new session
  • Fixed pinned background agents being re-prompted to "Continue from where you left off" after every auto-update
  • Fixed backgrounding the main turn spawning a phantom "general-purpose (resumed)" subagent that re-ran the main conversation
  • Fixed agent panel hiding sibling agents when viewing a subagent
  • Improved background agents: the launch result no longer instructs Claude to "end your response" — it keeps working on other tasks while the agent runs
  • … +3 more CLI changelog entries

Source: https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md#21193

Metadata

  • Time since 2.1.191 release: 1d 6h 54m 8s

  • Bundle file size: +102.8 kB (+0.4%)

  • Prompt files: +2 (+11.8%)

  • Prompt tokens: +198 (+0.9%)

Prompt token mix by kind:

  • tools: 68.2%→49.0%
  • system-reminder: 18.8%→38.1%
  • system-data: 6.0%→5.9%
  • skill: 3.0%→3.0%
  • system: 2.6%→2.6%
  • agent: 1.4%→1.4%

Files:

Other prompt changes

  1. Clarifies preserving exact indentation when editing files from the Read tool and updates the example schema to use a URL value.
  2. Documented grep as a ripgrep-based search tool and recommended to always use it for search tasks.

Links:

  1. b81d9e0/system-prompts/tool-description-edit.md#diff-d98c9d5750be0d3d005e4bf2b7153b7a85d658c33884e9c9ef520cd58aa2536cL3
  2. 88ec5c8/system-prompts/tool-description-grep.md#diff-98f084cc9ae739dcef3d2c75ea5cf55429dbb647973737f30693e59689d25d58R3