[Release] anthropics/claude-code v2.1.199: v2.1.199

Claude Code v2.1.199 is a maintenance release focused on reliability and user experience improvements. Key fixes address SSL certificate handling, streaming response preservation, subagent error reporting, background agent stability on Linux and macOS, and UI/UX issues like progress indicators and agent panel display. The release also improves retry logic for transient server errors and fixes several edge cases in session management and tool routing.

Key Points

• •Stacked slash-skill invocations now support up to 5 leading skills (e.g., `/skill-a /skill-b do XYZ`) instead of just the first
• •SSL/TLS certificate errors now fail immediately with actionable guidance instead of burning retries with generic messages
• •Streaming responses with mid-stream API errors are now preserved as partial output with an incomplete-response notice
• •Subagents now properly return partial work to parent agents when cut off by rate limits or server errors
• •Subagent API errors (e.g., usage limits) are now correctly reported to parent agents instead of being masked as successful results
• •Fixed Linux background-agent daemon self-termination issue occurring ~50 seconds after unclean shutdown with corrupted worker records
• •Background agents on macOS now cold-start correctly over SSH without 'Could not switch to audit session' errors
• •Transient 429 rate-limit errors are now automatically retried with backoff for subscribers; retry limits increased (300 default, no 15-cap)
• •Background job progress indicators no longer stall during long-running commands; memory-starved sessions now show specific low-memory errors
• •Fixed subagent visibility issues: idle agents collapse into expandable summary rows; agent panel no longer flaps between Working/Idle states

Workflow Diagram

Concepts