Skip to main content

Roo Code 3.44 Release Notes (2026-01-27)

This release improves worktree workflows, adds MCP configuration flexibility, and fixes context condensing and provider issues.

Roo Code v3.44 Release

Worktrees​

Worktrees are easier to work with in chat. The Worktree selector is more prominent, creating a worktree takes fewer steps, and the Create Worktree flow is clearer (including a native folder picker), so it’s faster to spin up an isolated branch/workspace and switch between worktrees while you work. (#10940)

📚 Documentation: See Worktrees for detailed usage.

Parallel tool calls (Experimental)​

Re-enables parallel tool calling (with new_task isolation safeguards) so you can use the experimental “Parallel tool calls” setting again without breaking task delegation workflows. (#11006)

QOL Improvements​

  • Makes subtasks easier to find and navigate by improving parent/child visibility across History and Chat (including clearer “back to parent” navigation), so you can move between related tasks faster. (#10864)
  • Lets you auto-approve all tools from a trusted MCP server by using alwaysAllow: ["*"], so you don’t have to list each tool name individually. (#10948)
  • Reduces token overhead in prompts by removing a duplicate MCP server/tools section from internal instructions, leaving more room for your conversation context. (#10895)
  • Improves Traditional Chinese (zh-TW) UI text for better clarity and consistency. (thanks PeterDaveHello!) (#10953)

Bug Fixes​

  • Fixes an issue where context condensing could accidentally pull in content that was already condensed earlier, which could reduce the effectiveness of long-conversation summaries. (#10985)
  • Fixes an issue where automatic context condensing could silently fail for VS Code LM API users when token counting returned 0 outside active requests, which could lead to unexpected context-limit errors. (thanks srulyt!) (#10983)
  • Fixes an issue where Roo didn’t record a successful truncation fallback when condensation failed, which could make Rewind restores unreliable after a condensing error. (#10984)
  • Fixes an issue where MCP tools with hyphens in their names could fail to resolve in native tool calling (for example when a provider/model rewrites - as _). (thanks hori-so!) (#10775)
  • Fixes an issue where tool calls could fail validation through AWS Bedrock when toolUseId exceeded Bedrock’s 64-character limit, improving reliability for longer tool-heavy sessions. (#10902)
  • Fixes an issue where Settings section headers could look transparent while scrolling, restoring an opaque background so the UI stays legible. (#10951)
  • Fixes a Fireworks provider type mismatch by removing unsupported model tool fields, keeping provider model metadata consistent and preventing breakage from schema changes. (#10937)
  • Fixes an issue where the new_task tool could miss creating a checkpoint before handing off, making task state more consistent and recoverable. (#10982)
  • Fixes an issue where leftover Power Steering experiment references could display raw translation keys in the UI. (#10980)
  • Fixes an issue where Roo could fail to index code in worktrees stored inside hidden directories (for example ~/.roo/worktrees/), which could break search and other codebase features in those worktrees. (#11009)

Provider Updates​

  • Adds the latest Fireworks models to the Fireworks provider’s model list. (thanks ThanhNguyxn!) (#10679)
  • Fixes an issue where tool-enabled chats could fail when using LiteLLM as a proxy to Amazon Bedrock due to Bedrock tool call ID validation limits. (#10990)
  • Sets default sampling parameters for the Cerebras zai-glm-4.7 model (temperature and top_p) so outputs are more consistent without extra configuration. (thanks sebastiand-cerebras!) (#10945)
  • Fixes an issue where local Ollama models could be incorrectly flagged as unavailable due to validation running against an empty router model list. (#10893)
  • Fixes an issue where tools could appear twice when using OpenAI Responses API-based providers, reducing duplicate tool output and making results easier to follow. (#11008)