|
| 1 | +--- |
| 2 | +description: Roo Code 3.36 introduces non-destructive context management, new debugging and UI controls, and a steady stream of reliability fixes and provider improvements. |
| 3 | +keywords: |
| 4 | + - roo code 3.36 |
| 5 | + - release notes |
| 6 | + - new features |
| 7 | + - bug fixes |
| 8 | + - provider updates |
| 9 | +image: /img/v3.36.0/v3.36.0.png |
| 10 | +--- |
| 11 | + |
| 12 | +# Roo Code 3.36 Release Notes (Combined) |
| 13 | + |
| 14 | +Roo Code 3.36 introduces non-destructive context management, new debugging and UI controls, and a steady stream of reliability fixes and provider improvements. |
| 15 | + |
| 16 | +<img src="/img/v3.36.0/v3.36.0.png" alt="Roo Code v3.36 Release" width="600" /> |
| 17 | + |
| 18 | +## Non-Destructive Context Management |
| 19 | + |
| 20 | +Context condensing and sliding window truncation now preserve your original messages internally rather than deleting them ([#9665](https://github.com/RooCodeInc/Roo-Code/pull/9665)). When you rewind to an earlier checkpoint, the full conversation history is restored automatically. |
| 21 | + |
| 22 | +## GPT-5.1 Codex Max Support |
| 23 | + |
| 24 | +Roo Code supports GPT-5.1 Codex Max, OpenAI’s long-horizon coding model, including model defaults for gpt-5.1 / gpt-5 / gpt-5-mini variants ([#9848](https://github.com/RooCodeInc/Roo-Code/pull/9848)). |
| 25 | + |
| 26 | +## Browser Screenshot Saving |
| 27 | + |
| 28 | +The browser tool can now save screenshots to a specified file path with a new `screenshot` action, so you can capture visual state during browser automation tasks ([#9963](https://github.com/RooCodeInc/Roo-Code/pull/9963)). |
| 29 | + |
| 30 | +## Extra-High Reasoning Effort |
| 31 | + |
| 32 | +If you use gpt-5.1-codex-max with the OpenAI provider, you can now select an “Extra High” reasoning effort level for maximum reasoning depth on complex tasks ([#9900](https://github.com/RooCodeInc/Roo-Code/pull/9900)). |
| 33 | + |
| 34 | +## OpenRouter Native Tools Default |
| 35 | + |
| 36 | +OpenRouter models that support native tools now use native tool calling by default, improving tool calling reliability without manual configuration ([#9878](https://github.com/RooCodeInc/Roo-Code/pull/9878)). |
| 37 | + |
| 38 | +## Error Details Modal |
| 39 | + |
| 40 | +Hover over error rows to reveal an info icon that opens a modal with full error details and a copy button ([#9985](https://github.com/RooCodeInc/Roo-Code/pull/9985)). |
| 41 | + |
| 42 | +## GPT-5.2 Model Support |
| 43 | + |
| 44 | +GPT-5.2 is available in the OpenAI provider and set as the default model ([#10024](https://github.com/RooCodeInc/Roo-Code/pull/10024)). |
| 45 | + |
| 46 | +## Enter Key Behavior Toggle |
| 47 | + |
| 48 | +You can now configure how Enter behaves in the chat input so it better fits multiline prompts and different input methods ([#10002](https://github.com/RooCodeInc/Roo-Code/pull/10002)). |
| 49 | + |
| 50 | +## QOL Improvements |
| 51 | + |
| 52 | +* **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 ([#9838](https://github.com/RooCodeInc/Roo-Code/pull/9838)) |
| 53 | +* **Smoother chat scroll**: Chat view maintains scroll position more reliably during streaming ([#8999](https://github.com/RooCodeInc/Roo-Code/pull/8999)) |
| 54 | +* **Clearer error messages**: More actionable errors with direct links to documentation ([#9777](https://github.com/RooCodeInc/Roo-Code/pull/9777)) |
| 55 | +* **Enter key behavior toggle**: Configure whether Enter sends or inserts a newline in chat input ([#10002](https://github.com/RooCodeInc/Roo-Code/pull/10002)) |
| 56 | +* **Unified context-management UX**: Real-time feedback for truncation notifications and condensation summaries ([#9795](https://github.com/RooCodeInc/Roo-Code/pull/9795)) |
| 57 | +* **Better OpenAI error messages**: Extracts more detail from API errors for easier troubleshooting ([#9639](https://github.com/RooCodeInc/Roo-Code/pull/9639)) |
| 58 | +* **Token counting optimization**: Removes separate API calls for token counting to improve performance ([#9884](https://github.com/RooCodeInc/Roo-Code/pull/9884)) |
| 59 | +* **Tool instructions decoupled from system prompts**: Tool-specific guidance is self-contained in tool descriptions ([#9784](https://github.com/RooCodeInc/Roo-Code/pull/9784)) |
| 60 | +* **Clearer auto-approve timing in follow-up suggestions**: Makes the auto-approve countdown harder to miss ([#10048](https://github.com/RooCodeInc/Roo-Code/pull/10048)) |
| 61 | + |
| 62 | +## Bug Fixes |
| 63 | + |
| 64 | +* **Write tool validation**: Avoids false positives where `write_to_file` rejected complete markdown files containing inline code comments like `# NEW:` or `// Step 1:` ([#9787](https://github.com/RooCodeInc/Roo-Code/pull/9787)) |
| 65 | +* **Download count display**: Fixes homepage download count precision for million-scale numbers ([#9807](https://github.com/RooCodeInc/Roo-Code/pull/9807)) |
| 66 | +* **Extension freeze prevention**: Avoids freezes when a model attempts to call a non-existent tool ([#9834](https://github.com/RooCodeInc/Roo-Code/pull/9834)) |
| 67 | +* **Checkpoint restore reliability**: Message history handling is consistent across rewind operations ([#9842](https://github.com/RooCodeInc/Roo-Code/pull/9842)) |
| 68 | +* **Context truncation fix**: Prevents cascading truncation loops by truncating only visible messages ([#9844](https://github.com/RooCodeInc/Roo-Code/pull/9844)) |
| 69 | +* **Reasoning models**: Models that require reasoning always receive valid reasoning-effort values ([#9836](https://github.com/RooCodeInc/Roo-Code/pull/9836)) |
| 70 | +* **Terminal input handling**: Inline terminal no longer hangs when commands require user input ([#9827](https://github.com/RooCodeInc/Roo-Code/pull/9827)) |
| 71 | +* **Large file safety**: Large file reads handle token budgets more safely ([#9843](https://github.com/RooCodeInc/Roo-Code/pull/9843)) |
| 72 | +* **Follow-up button styling**: Fixes overly rounded corners on follow-up suggestions ([#9829](https://github.com/RooCodeInc/Roo-Code/pull/9829)) |
| 73 | +* **Chutes provider fix**: Resolves model fetching errors by making schema validation more robust for optional fields ([#9854](https://github.com/RooCodeInc/Roo-Code/pull/9854)) |
| 74 | +* **Tool protocol selector**: Always shows the tool protocol selector for OpenAI-compatible providers ([#9966](https://github.com/RooCodeInc/Roo-Code/pull/9966)) |
| 75 | +* **apply_diff filtering**: Properly excludes apply_diff from native tools when diff is disabled ([#9920](https://github.com/RooCodeInc/Roo-Code/pull/9920)) |
| 76 | +* **API timeout handling**: Fixes a disabled timeout (set to 0) causing immediate request failures ([#9960](https://github.com/RooCodeInc/Roo-Code/pull/9960)) |
| 77 | +* **Reasoning effort dropdown**: Respects explicit supportsReasoningEffort values and fixes disable handling ([#9970](https://github.com/RooCodeInc/Roo-Code/pull/9970), [#9930](https://github.com/RooCodeInc/Roo-Code/pull/9930)) |
| 78 | +* **Actual error messages on retry**: Displays the provider’s error details instead of generic text ([#9954](https://github.com/RooCodeInc/Roo-Code/pull/9954)) |
| 79 | +* **Stream hanging fix**: Ensures finish_reason triggers tool_call_end events for multiple providers ([#9927](https://github.com/RooCodeInc/Roo-Code/pull/9927), [#9929](https://github.com/RooCodeInc/Roo-Code/pull/9929)) |
| 80 | +* **tool_result ID validation**: Validates and fixes tool_result IDs before requests to prevent provider rejections ([#9952](https://github.com/RooCodeInc/Roo-Code/pull/9952)) |
| 81 | +* **Suppressed internal error**: Fixes an internal “ask promise was ignored” error leaking to conversations ([#9914](https://github.com/RooCodeInc/Roo-Code/pull/9914)) |
| 82 | +* **Provider sanitization**: Fixes an infinite loop when using removed/invalid API providers ([#9869](https://github.com/RooCodeInc/Roo-Code/pull/9869)) |
| 83 | +* **Context icons theme**: Context-management icons now use foreground color to match VS Code themes ([#9912](https://github.com/RooCodeInc/Roo-Code/pull/9912)) |
| 84 | +* **Eval runs deletion**: Fixes a foreign key constraint preventing eval run deletions ([#9909](https://github.com/RooCodeInc/Roo-Code/pull/9909)) |
| 85 | +* **OpenAI-compatible timeout reliability**: Adds timeout handling to prevent indefinite hangs ([#9898](https://github.com/RooCodeInc/Roo-Code/pull/9898)) |
| 86 | +* **MCP tool streaming**: Fixes MCP tools failing with “unknown tool” errors due to premature clearing of internal streaming data ([#9993](https://github.com/RooCodeInc/Roo-Code/pull/9993)) |
| 87 | +* **TODO list display order**: TODO items display in execution order instead of being grouped by status ([#9991](https://github.com/RooCodeInc/Roo-Code/pull/9991)) |
| 88 | +* **Telemetry improvements**: Filters out 429 rate limit errors from API error telemetry for cleaner metrics ([#9987](https://github.com/RooCodeInc/Roo-Code/pull/9987)) |
| 89 | +* **Gemini stability**: Fixes reasoning loops and empty response errors ([#10007](https://github.com/RooCodeInc/Roo-Code/pull/10007)) |
| 90 | +* **Parallel tool execution**: Fixes “Expected toolResult blocks at messages” errors during parallel tool use ([#10015](https://github.com/RooCodeInc/Roo-Code/pull/10015)) |
| 91 | +* **tool_result ID mismatch**: Fixes ToolResultIdMismatchError when history has orphaned tool_result blocks ([#10027](https://github.com/RooCodeInc/Roo-Code/pull/10027)) |
| 92 | +* **Parallel tool calls fix**: Preserves tool_use blocks in summaries during context condensation to avoid API errors with parallel tool calling ([#9714](https://github.com/RooCodeInc/Roo-Code/pull/9714)) |
| 93 | +* **Navigation button wrapping**: Prevents navigation buttons from wrapping on smaller screens ([#9721](https://github.com/RooCodeInc/Roo-Code/pull/9721)) |
| 94 | +* **Task delegation tool flush**: Ensures pending tool results are flushed before delegating tasks to avoid provider 400 errors ([#9726](https://github.com/RooCodeInc/Roo-Code/pull/9726)) |
| 95 | +* **Malformed tool call handling**: Prevents the extension from hanging indefinitely on malformed tool calls by validating and reporting missing parameters ([#9758](https://github.com/RooCodeInc/Roo-Code/pull/9758)) |
| 96 | +* **Auto-approval stops when you start typing**: Fixes an issue where an auto-approve timer could still fire after you began writing a response ([#9937](https://github.com/RooCodeInc/Roo-Code/pull/9937)) |
| 97 | +* **More actionable OpenRouter error messages**: Surfaces upstream error details when available ([#10039](https://github.com/RooCodeInc/Roo-Code/pull/10039)) |
| 98 | +* **LiteLLM tool protocol dropdown always appears**: Restores the tool protocol dropdown in Advanced settings even when model metadata isn’t available yet ([#10053](https://github.com/RooCodeInc/Roo-Code/pull/10053)) |
| 99 | +* **MCP tool calls work with stricter providers**: Avoids failures caused by special characters in MCP server/tool names by sanitizing names and using an unambiguous `mcp--server--tool` ID format ([#10054](https://github.com/RooCodeInc/Roo-Code/pull/10054)) |
| 100 | + |
| 101 | +## Misc Improvements |
| 102 | + |
| 103 | +* **Evals UI enhancements**: Adds better filtering, bulk delete actions, tool column consolidation, and run notes ([#9837](https://github.com/RooCodeInc/Roo-Code/pull/9837)) |
| 104 | +* **Multi-model evals launch**: Launches identical test runs across multiple models with automatic staggering ([#9845](https://github.com/RooCodeInc/Roo-Code/pull/9845)) |
| 105 | +* **New pricing page**: Updates the website pricing page with clearer feature explanations ([#9821](https://github.com/RooCodeInc/Roo-Code/pull/9821)) |
| 106 | +* **Announcement UI updates**: Improves announcement visuals with updated social icons and GitHub stars CTA ([#9945](https://github.com/RooCodeInc/Roo-Code/pull/9945)) |
| 107 | +* **Improved error logging**: Adds better context for parseToolCall exceptions and cloud job errors ([#9857](https://github.com/RooCodeInc/Roo-Code/pull/9857), [#9924](https://github.com/RooCodeInc/Roo-Code/pull/9924)) |
| 108 | +* **search_replace native tool**: Adds a tool for single-replacement file operations with precise targeting via unique text matching ([#9918](https://github.com/RooCodeInc/Roo-Code/pull/9918)) |
| 109 | +* **Versioned settings support**: Adds internal infrastructure for API-side versioning of model settings with minimum plugin version gating ([#9934](https://github.com/RooCodeInc/Roo-Code/pull/9934)) |
| 110 | +* **OpenRouter telemetry**: Adds API error telemetry for better diagnostics ([#9953](https://github.com/RooCodeInc/Roo-Code/pull/9953)) |
| 111 | +* **Evals streaming stats**: Tool usage stats stream in real time with token usage throttling ([#9926](https://github.com/RooCodeInc/Roo-Code/pull/9926)) |
| 112 | +* **Tool consolidation**: Removes the deprecated `insert_content` tool (use apply_diff or write_to_file) ([#9751](https://github.com/RooCodeInc/Roo-Code/pull/9751)) |
| 113 | +* **Experimental settings**: Temporarily disables the parallel tool calls experiment while improvements are in progress ([#9798](https://github.com/RooCodeInc/Roo-Code/pull/9798)) |
| 114 | +* **Infrastructure**: Updates Next.js dependencies for web applications ([#9799](https://github.com/RooCodeInc/Roo-Code/pull/9799)) |
| 115 | +* **Removed deprecated tool**: Removes the deprecated `list_code_definition_names` tool ([#10005](https://github.com/RooCodeInc/Roo-Code/pull/10005)) |
| 116 | +* **Tool aliases for model-specific tool naming**: Adds support for alternative tool names so different models can call the same tool using the naming they expect ([#9989](https://github.com/RooCodeInc/Roo-Code/pull/9989)) |
| 117 | +* **Workspace task visibility controls for organizations**: Adds an org-level setting for how visible Roo Code Cloud “extension tasks” are across the workspace ([#10020](https://github.com/RooCodeInc/Roo-Code/pull/10020)) |
| 118 | + |
| 119 | +## Provider Updates |
| 120 | + |
| 121 | +* **Reasoning details support (Roo provider)**: Displays reasoning details from supported models ([#9796](https://github.com/RooCodeInc/Roo-Code/pull/9796)) |
| 122 | +* **Native tools default (Roo provider)**: Roo provider models default to native tool protocol ([#9811](https://github.com/RooCodeInc/Roo-Code/pull/9811)) |
| 123 | +* **MiniMax search_and_replace**: MiniMax M2 uses search_and_replace for more reliable file edits ([#9780](https://github.com/RooCodeInc/Roo-Code/pull/9780)) |
| 124 | +* **Cerebras token optimization**: Prevents premature rate limiting and cleans up deprecated models ([#9804](https://github.com/RooCodeInc/Roo-Code/pull/9804)) |
| 125 | +* **Vercel AI Gateway**: More reliable model fetching when pricing data is incomplete ([#9791](https://github.com/RooCodeInc/Roo-Code/pull/9791)) |
| 126 | +* **Dynamic model settings (Roo provider)**: Roo models receive configuration dynamically from the API ([#9852](https://github.com/RooCodeInc/Roo-Code/pull/9852)) |
| 127 | +* **Optimized GPT-5 tool configuration**: GPT-5.x, GPT-5.1.x, and GPT-4.1 use only apply_patch for file edits ([#9853](https://github.com/RooCodeInc/Roo-Code/pull/9853)) |
| 128 | +* **DeepSeek V3.2**: Updates to V3.2 with a price reduction, native tools by default, and 8K max output ([#9962](https://github.com/RooCodeInc/Roo-Code/pull/9962)) |
| 129 | +* **xAI models catalog**: Corrects context windows, adds image support for grok-3/grok-3-mini, and removes deprecated models ([#9872](https://github.com/RooCodeInc/Roo-Code/pull/9872)) |
| 130 | +* **xAI tool preferences**: Configures xAI models to use search_replace for better file editing compatibility ([#9923](https://github.com/RooCodeInc/Roo-Code/pull/9923)) |
| 131 | +* **DeepSeek V3.2 for Baseten**: Adds DeepSeek V3.2 model support ([#9861](https://github.com/RooCodeInc/Roo-Code/pull/9861)) |
| 132 | +* **Baseten model tweaks**: Improves maxTokens limits and native tools support for stability ([#9866](https://github.com/RooCodeInc/Roo-Code/pull/9866)) |
| 133 | +* **Bedrock models**: Adds Kimi, MiniMax, and Qwen model configurations ([#9905](https://github.com/RooCodeInc/Roo-Code/pull/9905)) |
| 134 | +* **Z.ai endpoint options**: Adds endpoint options for users on API billing instead of the Coding plan ([#9894](https://github.com/RooCodeInc/Roo-Code/pull/9894)) |
0 commit comments