|
1 | 1 | # Roo Code Changelog |
2 | 2 |
|
| 3 | +## [3.14.3] - 2025-04-25 |
| 4 | + |
| 5 | +- Add Boomerang Orchestrator as a built-in mode |
| 6 | +- Improve home screen UI |
| 7 | +- Make token count estimation more efficient to reduce gray screens |
| 8 | +- Revert change to automatically close files after edit until we figure out how to make it work well with diagnostics |
| 9 | +- Clean up settings data model |
| 10 | +- Omit reasoning params for non-reasoning models |
| 11 | +- Clearer documentation for adding settings (thanks @shariqriazz!) |
| 12 | +- Fix word wrapping in Roo message title (thanks @zhangtony239!) |
| 13 | +- Update default model id for Unbound from claude 3.5 to 3.7 (thanks @pugazhendhi-m!) |
| 14 | + |
| 15 | +## [3.14.2] - 2025-04-24 |
| 16 | + |
| 17 | +- Enable prompt caching for Gemini (with some improvements) |
| 18 | +- Allow users to turn prompt caching on / off for Gemini 2.5 on OpenRouter |
| 19 | +- Compress terminal output with backspace characters (thanks @KJ7LNW) |
| 20 | +- Add Russian language (Спасибо @asychin) |
| 21 | + |
| 22 | +## [3.14.1] - 2025-04-24 |
| 23 | + |
| 24 | +- Disable Gemini caching while we investigate issues reported by the community. |
| 25 | + |
| 26 | +## [3.14.0] - 2025-04-23 |
| 27 | + |
| 28 | +- Add prompt caching for `gemini-2.5-pro-preview-03-25` in the Gemini provider (Vertex and OpenRouter coming soon!) |
| 29 | +- Improve the search_and_replace and insert_content tools and bring them out of experimental, and deprecate append_to_file (thanks @samhvw8!) |
| 30 | +- Use material icons for files and folders in mentions (thanks @elianiva!) |
| 31 | +- Make the list_files tool more efficient and smarter about excluding directories like .git/ |
| 32 | +- Fix file drag and drop on Windows and when using SSH tunnels (thanks @NyxJae!) |
| 33 | +- Correctly revert changes and suggest alternative tools when write_to_file fails on a missing line count |
| 34 | +- Allow interpolation of `workspace`, `mode`, `language`, `shell`, and `operatingSystem` into custom system prompt overrides (thanks @daniel-lxs!) |
| 35 | +- Fix interpolation bug in the “add to context” code action (thanks @elianiva!) |
| 36 | +- Preserve editor state and prevent tab unpinning during diffs (thanks @seedlord!) |
| 37 | +- Improvements to icon rendering on Linux (thanks @elianiva!) |
| 38 | +- Improvements to Requesty model list fetching (thanks @dtrugman!) |
| 39 | +- Fix user feedback not being added to conversation history in API error state, redundant ‘TASK RESUMPTION’ prompts, and error messages not showing after cancelling API requests (thanks @System233!) |
| 40 | +- Track tool use errors in evals |
| 41 | +- Fix MCP hub error when dragging extension to another sidebar |
| 42 | +- Improve display of long MCP tool arguments |
| 43 | +- Fix redundant ‘TASK RESUMPTION’ prompts (thanks @System233!) |
| 44 | +- Fix bug opening files when editor has no workspace root |
| 45 | +- Make the VS Code LM provider show the correct model information (thanks @QuinsZouls!) |
| 46 | +- Fixes to make the focusInput command more reliable (thanks @hongzio!) |
| 47 | +- Better handling of aftercursor content in context mentions (thanks @elianiva!) |
| 48 | +- Support injecting environment variables in MCP config (thanks @NamesMT!) |
| 49 | +- Better handling of FakeAI “controller” object (thanks @wkordalski) |
| 50 | +- Remove unnecessary calculation from VS Code LM provider (thanks @d-oit!) |
| 51 | +- Allow Amazon Bedrock Marketplace ARNs (thanks @mlopezr!) |
| 52 | +- Give better loading feedback on chat rows (thanks @elianiva!) |
| 53 | +- Performance improvements to task size calculations |
| 54 | +- Don’t immediately show a model ID error when changing API providers |
| 55 | +- Fix apply_diff edge cases |
| 56 | +- Use a more sensible task export icon |
| 57 | +- Use path aliases in webview source files |
| 58 | +- Display a warning when the system prompt is overridden |
| 59 | +- Better progress indicator for apply_diff tools (thanks @qdaxb!) |
| 60 | +- Fix terminal carriage return handling for correct progress bar display (thanks @Yikai-Liao!) |
| 61 | + |
| 62 | +## [3.13.2] - 2025-04-18 |
| 63 | + |
| 64 | +- Allow custom URLs for Gemini provider |
| 65 | + |
| 66 | +## [3.13.1] - 2025-04-18 |
| 67 | + |
| 68 | +- Support Gemini 2.5 Flash thinking mode (thanks @monotykamary) |
| 69 | +- Make auto-approval toggle on/off states more obvious (thanks @sachasayan) |
| 70 | +- Add telemetry for shell integration errors |
| 71 | +- Fix the path of files dragging into the chat textarea on Windows (thanks @NyxJae) |
| 72 | + |
| 73 | +## [3.13.0] - 2025-04-17 |
| 74 | + |
| 75 | +- UI improvements to task header, chat view, history preview, and welcome view (thanks @sachasayan!) |
| 76 | +- Add append_to_file tool for appending content to files (thanks @samhvw8!) |
| 77 | +- Add Gemini 2.5 Flash Preview to Gemini and Vertex providers (thanks @nbihan-mediware!) |
| 78 | +- Fix image support in Bedrock (thanks @Smartsheet-JB-Brown!) |
| 79 | +- Make diff edits more resilient to models passing in incorrect parameters |
| 80 | + |
| 81 | +## [3.12.3] - 2025-04-17 |
| 82 | + |
| 83 | +- Fix character escaping issues in Gemini diff edits |
| 84 | +- Support dragging and dropping tabs into the chat box (thanks @NyxJae!) |
| 85 | +- Make sure slash commands only fire at the beginning of the chat box (thanks @logosstone!) |
| 86 | + |
| 87 | +## [3.12.2] - 2025-04-16 |
| 88 | + |
| 89 | +- Add OpenAI o3 & 4o-mini (thanks @PeterDaveHello!) |
| 90 | +- Improve file/folder context mention UI (thanks @elianiva!) |
| 91 | +- Improve diff error telemetry |
| 92 | + |
| 93 | +## [3.12.1] - 2025-04-16 |
| 94 | + |
| 95 | +- Bugfix to Edit button visibility in the select dropdowns |
| 96 | + |
| 97 | +## [3.12.0] - 2025-04-15 |
| 98 | + |
| 99 | +- Add xAI provider and expose reasoning effort options for Grok on OpenRouter (thanks Cline!) |
| 100 | +- Make diff editing config per-profile and improve pre-diff string normalization |
| 101 | +- Make checkpoints faster and more reliable |
| 102 | +- Add a search bar to mode and profile select dropdowns (thanks @samhvw8!) |
| 103 | +- Add telemetry for code action usage, prompt enhancement usage, and consecutive mistake errors |
| 104 | +- Suppress zero cost values in the task header (thanks @do-it!) |
| 105 | +- Make JSON parsing safer to avoid crashing the webview on bad input |
| 106 | +- Allow users to bind a keyboard shortcut for accepting suggestions or input in the chat view (thanks @axkirillov!) |
| 107 | + |
3 | 108 | ## [3.11.17] - 2025-04-14 |
4 | 109 |
|
5 | 110 | - Improvements to OpenAI cache reporting and cost estimates (thanks @monotykamary and Cline!) |
|
0 commit comments