Skip to main content

Roo Code 3.36.1 Release Notes (2025-12-04)

This release adds GPT-5.1 Codex Max support, enables symlinks for slash commands, improves message management with better checkpoint restore, and fixes several stability issues.

Roo Code v3.36.1 Release

GPT-5.1 Codex Max Support​

Roo Code now supports GPT-5.1 Codex Max (#9848), OpenAI's most intelligent coding model optimized for long-horizon, agentic coding tasks. This release also adds model defaults for gpt-5.1, gpt-5, and gpt-5-mini variants with optimized configurations.

📚 Documentation: See OpenAI Provider for configuration details.

QOL Improvements​

  • Symlink support for slash commands: Share and organize commands across projects using symlinks for individual files or directories, with command names derived from symlink names for easy aliasing (#9838)
  • Smoother chat scroll: Chat view maintains scroll position more reliably during streaming, eliminating disruptive jumps (#8999)
  • Improved error messages: Clearer, more actionable error messages with proper attribution and direct links to documentation (#9777)

Bug Fixes​

  • Extension freeze prevention: The extension no longer freezes when a model attempts to call a non-existent tool (thanks daniel-lxs!) (#9834)
  • Checkpoint restore reliability: MessageManager layer ensures consistent message history handling across all rewind operations (#9842)
  • Context truncation fix: Prevent cascading truncation loops by only truncating visible messages (#9844)
  • Reasoning models: Models that require reasoning now always receive valid reasoning effort values (#9836)
  • Terminal input handling: Inline terminal no longer hangs when commands require user input (#9827)
  • Large file safety: Safer large file reads with proper token budget accounting for model output (#9843)
  • Follow-up button styling: Fixed overly rounded corners on follow-up question suggestions (#9829)

Misc Improvements​

  • Evals UI enhancements: Added filtering by timeframe/model/provider, bulk delete actions, tool column consolidation, and run notes (#9837)
  • Multi-model evals launch: Launch identical test runs across multiple models with automatic staggering (#9845)
  • New pricing page: Updated website pricing page with clearer feature explanations (#9821)