Claude Code v2.1.296 Release - Workflow Agent Model Configuration and Large File Reading Features Added

Claude Code has released v2.1.296 update. Key additions include workflow agent-specific model configuration (CLAUDE_CODE_WORKFLOW_SUBAGENT_MODEL), bulk reading

Claude Code · Release v2.1.296 · 2026-10-09

Claude Code has released v2.1.296 update. Key additions include workflow agent-specific model configuration (CLAUDE_CODE_WORKFLOW_SUBAGENT_MODEL), bulk reading option for large text files (allow_large), and overload retry maximum delay settings. Numerous stability improvements were made including fixes for managed configuration PreToolUse/PostToolUse hook handling errors, token counting issues, and plugin session start bugs. Self-hosted runner git operation failures and workflow script result truncation issues were also resolved.

weltiq perspective: Improved stability and flexibility of AI agent-based automation tools reduces the likelihood of errors when weltiq users integrate external code tools during backtesting and strategy execution.

Release Notes Excerpt

What's changed Added a code key to the Claude apps gateway's managed.policies[] : the same settings as cli , also applied in Claude Desktop's Code tab; beside desktop , it turns on Claude Desktop's gateway mode Added autoCompactWindow to subagent frontmatter and --agents definitions, so a subagent can auto-compact earlier than the main conversation's window Added CLAUDE_CODE_WORKFLOW_SUBAGENT_MODEL to run every workflow agent on one model while other subagents keep theirs Added CLAUDE_CODE_OVERLOADED_RETRY_MAX_DELAY_MS environment variable to set a longer maximum delay for the backoff when retrying an overloaded (529) request Added a note in /plugin on a plugin whose hooks are left out becau…

Source: https://github.com/anthropics/claude-code/releases/tag/v2.1.296