Claude Code v2.1.280: Opus 5.5 Model Added and Stability Improvements

Claude Code has added the new default model Claude Opus 5.5 (1M token context, $4/$20/Mtok). Fixed infinite retry bug when safety checks are rejected in auto mo

Claude Code · Release v2.1.280 · 2026-09-22

Claude Code has added the new default model Claude Opus 5.5 (1M token context, $4/$20/Mtok). Fixed infinite retry bug when safety checks are rejected in auto mode and resolved symbolic link path write judgment errors. Added environment variable to set MCP tool description length limits and improved mouse support and keyboard input handling.

weltiq perspective: With longer context and stable auto-execution mode, developers can build AI agent-based trading tools more effectively, benefiting weltiq's backtest and automated execution pipeline construction.

Release Notes Excerpt

What's changed Added Claude Opus 5.5 ( claude-opus-5-5 ), now the default Opus model — 1M context, $4/$20 per Mtok with $0.20/Mtok cache reads Added mouse support to more lists in fullscreen mode: the wheel scrolls the /skills list, and a skill's state options in /plugin can be clicked Added CLAUDE_CODE_MAX_MCP_DESCRIPTION_LENGTH to change the 2,048-character cap on MCP tool descriptions and server instructions for every MCP server in the session Added hook output sizes and the number of oversized outputs saved to a file to the hook_execution_complete OpenTelemetry event Fixed writes through a symlinked path being judged by their in-tree spelling: the prompt names where the write lands, and …

Source: https://github.com/anthropics/claude-code/releases/tag/v2.1.280