DAILY RELEASE TRACKER
Anthropic’s terminal-native coding agent and developer workflow tool.
The v2.1.308 release for Claude Code focuses exclusively on bug fixes and reliability improvements, addressing underlying stability issues without introducing new features. This update ensures a more consistent and dependable development environment for technical teams relying on the tool for automated coding tasks.
This update introduces granular OpenTelemetry safety instrumentation and an explicit API versioning command, enhancing observability for AI product teams while resolving critical stability issues with MCP server authentication, retry logic, and context window handling on macOS and Linux.
This update resolves a critical authentication regression where the `claude -p` command failed with a "Not logged in" error in cloud sessions lacking explicit environment variables, ensuring reliable CLI access. It also corrects a logic flaw in sessions with many MCP tools where API errors were incorrectly masked as "prompt is too long" and triggered unnecessary compaction, thereby preserving accurate error diagnostics and optimal context window usage.
This update resolves critical regressions in tool search handling for Vertex and Bedrock providers, ensuring requests no longer fail when gateways reject tool search capabilities. It also fixes configuration persistence issues for `/effort` and `/autocompact` settings when model names contain case variations or `[1m]` tags, and corrects worktree initialization in bare repositories to prevent subagent cleanup failures.
This update enhances modding and tooling control by introducing explicit read-only flags and human-decision requirements for tool checks, while improving transparency through detailed hook-skipping transcripts. It also expands auto-mode setup to more environments and fixes critical stability issues regarding sandboxed commands, API key policy refreshing, and transcript garbage collection.
The v2.1.303 release for Claude Code focuses on bug fixes and reliability improvements, addressing underlying stability issues within the CLI environment. This update ensures more consistent execution for developers and operators relying on automated coding workflows, reducing unexpected failures during integration or deployment tasks.
This update introduces granular control over client-side auto-mode model selection via the `CLAUDE_CODE_AUTO_MODE_MODEL` environment variable and provides explicit configuration toggles for idle compaction, enhancing operational flexibility. It also resolves critical stability issues, including memory exhaustion risks in the Read tool, prompt cache misses, and configuration write race conditions, while deprecating standalone Claude Design access.
This update enhances operational resilience by introducing a configurable timeout for unattended retries to prevent indefinite waits on usage limits, while also improving observability through detailed OpenTelemetry tracing for classifier decisions. Additionally, it refines developer ergonomics with vim mode enhancements and multi-line input fields, alongside critical bug fixes for worktree safety and API error handling.