Skip to content

Commit 9b08411

Browse files
hannesrudolphhannesrudolph
committed
Add release notes for v3.23.17
- Created release notes file (v3.23.17.mdx) - Updated index.md with new version entry - Updated sidebars.ts with new version entry - Updated combined release notes (v3.23.mdx) with v3.23.17 changes
1 parent a448afc commit 9b08411

File tree

5 files changed

+341
-0
lines changed

5 files changed

+341
-0
lines changed

‎docs/update-notes/index.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -19,6 +19,7 @@ This section contains notes about recent updates to Roo Code, listed by version
1919

2020
## Version 3.23
2121

22+
* [3.23.17](/update-notes/v3.23.17) (2025-07-22)
2223
* [3.23.16](/update-notes/v3.23.16) (2025-07-19)
2324
* [3.23.15](/update-notes/v3.23.15) (2025-07-18)
2425
* [3.23.14](/update-notes/v3.23.14) (2025-07-17)

‎docs/update-notes/temp_pr_analysis.md‎

Lines changed: 267 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,267 @@
1+
# PR Analysis for Release Notes v3.23.17
2+
3+
## PR #6091: Release v3.23.17
4+
5+
**Author**: app/roomote
6+
**Linked Issues**: None (This is a release preparation PR)
7+
**Category**: Release (Multiple bug fixes, features, and improvements)
8+
**User Impact**: This release brings numerous improvements including better error handling, new AI model support, enhanced UI features, and various bug fixes that improve overall stability and user experience.
9+
**Contributors**: app/roomote
10+
11+
### Detailed Changes:
12+
13+
**Bug Fixes:**
14+
- **Context Management**: Context condensing prompt moved to proper Prompts section for better organization
15+
- **Terminal Output**: Added character limit to prevent context explosion from excessive terminal output
16+
- **Mode Export**: Fixed global mode export to properly include rules files
17+
- **UI Controls**: Export, share, and copy buttons now remain enabled during API operations
18+
- **File Organization**: Symlinked rules files now sort correctly by symlink names
19+
- **Testing**: Added missing run parameter to vitest command in rules
20+
- **Error Messages**: Better distinction between user cancellations and API failures
21+
- **Cloud Sessions**: More specific error codes for session issues
22+
- **Evals**: Fixed issues introduced by PR #5865
23+
- **Apply Diff**: Added case sensitivity reminder in error messages
24+
25+
**New Features:**
26+
- **File Navigation**: Jump icon for quickly accessing newly created files
27+
- **Evals Configuration**: Configurable timeout slider (5-10 minutes) for evaluation runs
28+
- **MCP Optimization**: Auto-omits MCP content when no servers are configured
29+
- **Provider Support**: Added Mistral embedding provider and Moonshot provider
30+
- **Model Support**: Added Llama-4-maverick to Vertex AI and Qwen3-235B-A22B-Instruct-2507 to Chutes AI
31+
- **Settings UI**: Todo list tool enable checkbox in provider advanced settings
32+
33+
**Improvements:**
34+
- **Token Management**: Enhanced max_tokens fallback logic in sliding window
35+
- **Documentation**: Clarified when to use update_todo_list tool
36+
37+
---
38+
39+
## PR #6052: feat: add Qwen/Qwen3-235B-A22B-Instruct-2507 model to Chutes AI provider
40+
41+
**Author**: app/roomote
42+
**Linked Issues**: #6050 (reporter: apple-techie)
43+
**Category**: Provider Update
44+
**User Impact**: Adds support for the Qwen/Qwen3-235B-A22B-Instruct-2507 model with a massive 262,144 token context window, enabling users to process extremely large documents, entire codebases, and lengthy technical documentation in a single request without breaking them into chunks.
45+
**Contributors**: apple-techie
46+
47+
---
48+
49+
## PR #6076: fix: add case sensitivity mention to suggested fixes in apply_diff error message
50+
51+
**Author**: app/roomote
52+
**Linked Issues**: #4731 (reporter: maskelihileci)
53+
**Category**: QOL Improvement
54+
**User Impact**: Clearer error messages when apply_diff fails due to case sensitivity mismatches. The error message now explicitly mentions that case sensitivity matters when matching content, helping users understand why their edits might fail.
55+
**Contributors**: maskelihileci
56+
57+
---
58+
59+
## PR #5738: Add jump icon for newly created files
60+
61+
**Author**: app/roomote
62+
**Linked Issues**: #5736 (reporter: mkdir700)
63+
**Category**: QOL Improvement
64+
**User Impact**: Users can now quickly open newly created files by clicking the jump icon, providing a consistent experience with edited files
65+
**Contributors**: app/roomote, mkdir700
66+
67+
---
68+
69+
## PR #6032: Add todo list tool enable checkbox to provider advanced settings
70+
71+
**Author**: app/roomote
72+
**Linked Issues**: None
73+
**Category**: QOL Improvement
74+
**User Impact**: Users can now control whether the todo list tool is available to the assistant, providing more flexibility in managing which tools are enabled for their workflow
75+
**Contributors**: app/roomote
76+
77+
---
78+
79+
## PR #5903: fix: sort symlinked rules files by symlink names, not target names
80+
81+
**Author**: app/roomote
82+
**Linked Issues**: #4131 (reporter: dsent)
83+
**Category**: Bug Fix
84+
**User Impact**: Custom rules files using symlinks now load in the correct alphabetical order. Users can reliably control the order of their rules files using numbered symlinks (e.g., 01-database.yaml, 02-api.yaml), making it easier to organize and prioritize custom instructions.
85+
**Contributors**: app/roomote, dsent
86+
87+
---
88+
89+
## PR #6025: fix: properly distinguish between user cancellations and API failures
90+
91+
**Author**: app/roomote
92+
**Linked Issues**: #5427 (reporter: MDean-Slalom)
93+
**Category**: Bug Fix
94+
**User Impact**: Fixed an issue where all API failures were incorrectly shown as "API Request Cancelled". Now users will see the correct error messages when API requests fail, making it easier to diagnose and resolve issues.
95+
**Contributors**: MDean-Slalom, daniel-lxs
96+
97+
---
98+
99+
## PR #5889: feat: auto-omit MCP content when no servers are configured
100+
101+
**Author**: app/roomote
102+
**Linked Issues**: None
103+
**Category**: QOL Improvement
104+
**User Impact**: Cleaner system prompts that automatically exclude MCP-related instructions when you're not using MCP servers, reducing unnecessary verbosity
105+
**Contributors**: app/roomote
106+
107+
---
108+
109+
## PR #5865: feat: add configurable timeout for evals (5-10 min)
110+
111+
**Author**: app/roomote
112+
**Linked Issues**: None
113+
**Category**: Feature
114+
**User Impact**: Users can now configure evaluation run timeouts between 5-10 minutes using a slider interface, allowing longer-running evaluations to complete without timing out.
115+
**Contributors**: app/roomote
116+
117+
---
118+
119+
## PR #5777: fix: add character limit to prevent terminal output context explosion
120+
121+
**Author**: app/roomote
122+
**Linked Issues**: #5775 (reporter: robottwo)
123+
**Category**: Bug Fix
124+
**User Impact**: Fixed a critical issue where terminal commands outputting extremely long lines without newlines would cause context window explosion and crashes. Roo Code now properly limits terminal output by character count (default 100,000 characters) in addition to line count, preventing crashes and ensuring stable operation even with commands that produce massive single-line outputs.
125+
**Contributors**: app/roomote, robottwo
126+
127+
---
128+
129+
## PR #6090: feat: add merge-resolver mode for intelligent conflict resolution
130+
131+
**Author**: daniel-lxs
132+
**Linked Issues**: None
133+
**Category**: Major Feature
134+
**User Impact**: Automatically resolve Git merge conflicts intelligently using commit history analysis. The new merge-resolver mode makes smart decisions about which changes to keep, prioritizing bugfixes, combining improvements, and preserving tests. This saves time and reduces errors when dealing with complex merge conflicts.
135+
**Contributors**: daniel-lxs
136+
137+
---
138+
139+
## PR #6019: fix: add bedrock to ANTHROPIC_STYLE_PROVIDERS and restore vertex Claude model checking
140+
141+
**Author**: daniel-lxs
142+
**Linked Issues**: #6004 (reporter: fedorbass)
143+
**Category**: Bug Fix
144+
**User Impact**: AWS Bedrock users will now see accurate token counts and won't experience false context window exceeded errors or issues with the chat history condenser.
145+
**Contributors**: daniel-lxs, fedorbass
146+
147+
---
148+
149+
## PR #5920: fix: add Git installation check for checkpoints feature (#3109)
150+
151+
**Author**: MuriloFP
152+
**Linked Issues**: #3109 (reporter: hannesrudolph)
153+
**Category**: QOL Improvement
154+
**User Impact**: Users now receive immediate feedback when Git is not installed, preventing silent failures of the checkpoints feature. A clear warning notification appears with a "Learn More" button that directs to the Git download page, helping users quickly resolve the issue.
155+
**Contributors**: MuriloFP
156+
157+
---
158+
159+
## PR #6023: feat: add llama-4-maverick model to Vertex AI provider (#5808)
160+
161+
**Author**: MuriloFP
162+
**Linked Issues**: #5808 (reporter: konstantinosbotonakis)
163+
**Category**: Provider Updates
164+
**User Impact**: Users can now select and use the Llama 4 Maverick model (with 128K context) when using Google Vertex AI as their provider in the US-EAST5 region
165+
**Contributors**: MuriloFP, konstantinosbotonakis
166+
167+
---
168+
169+
## PR #5849: fix: enable export, share, and copy buttons during API operations (#5324)
170+
171+
**Author**: MuriloFP
172+
**Linked Issues**: #5324 (reporter: KJ7LNW)
173+
**Category**: Bug Fix
174+
**User Impact**: Fixed a regression that prevented users from exporting task history, sharing tasks, or copying content while the API was running. These actions are now available during model operations, restoring an important debugging workflow that was broken between versions 3.16.5 and 3.22.5.
175+
**Contributors**: MuriloFP, KJ7LNW
176+
177+
---
178+
179+
## PR #5926: docs: clarify when to use update_todo_list tool
180+
181+
**Author**: hannesrudolph
182+
**Linked Issues**: None
183+
**Category**: Documentation Update
184+
**User Impact**: Clarifies that the todo list tool should be used for complicated tasks, not just multi-step tasks, helping users better understand when to leverage this feature for task management.
185+
**Contributors**: hannesrudolph
186+
187+
---
188+
189+
## PR #5837: fix: resolve global mode export not including rules files (#5834)
190+
191+
**Author**: hannesrudolph
192+
**Linked Issues**: #5834 (reporter: hannesrudolph)
193+
**Category**: Bug Fix
194+
**User Impact**: Fixed an issue where exporting global custom modes did not include their associated rules files. Now when you export a global mode, all rules files from the global configuration directory are properly included in the export, making it easier to share and backup your custom mode configurations.
195+
**Contributors**: hannesrudolph (excluded from release notes per rules)
196+
197+
---
198+
199+
## PR #6065: Fix evals; broken by #5865
200+
201+
**Author**: cte
202+
**Linked Issues**: None
203+
**Category**: Bug Fix
204+
**User Impact**: Fixed the evaluation system that was broken by a previous update, restoring proper timeout handling and service startup functionality
205+
**Contributors**: cte
206+
207+
---
208+
209+
## PR #6051: Bugfix: Cloud: be more specific about session error codes
210+
211+
**Author**: jr
212+
**Linked Issues**: None
213+
**Category**: Bug Fix
214+
**User Impact**: Prevents unnecessary logouts during service outages or rate limiting. Users will maintain their session during temporary service issues instead of being forced to log in again.
215+
**Contributors**: jr
216+
217+
---
218+
219+
## PR #6046: feat: add moonshot provider
220+
221+
**Author**: mrubens
222+
**Linked Issues**: #5869 (reporter: CellenLee)
223+
**Category**: Provider Updates
224+
**User Impact**: Adds support for Moonshot AI models, giving users another AI provider option to choose from in Roo Code
225+
**Contributors**: mrubens, CellenLee
226+
227+
---
228+
229+
## PR #5993: Update the max_tokens fallback logic in the sliding window
230+
231+
**Author**: mrubens
232+
**Linked Issues**: None
233+
**Category**: Misc Improvements
234+
**User Impact**: More predictable context handling with a consistent 8192 token fallback instead of variable 20% calculations, ensuring reliable behavior when working with large conversations.
235+
**Contributors**: mrubens
236+
237+
---
238+
239+
## PR #5991: fix: add run parameter to vitest command in rules
240+
241+
**Author**: KJ7LNW
242+
**Linked Issues**: #5992 (reporter: KJ7LNW)
243+
**Category**: Bug Fix
244+
**User Impact**: Fixed an issue where vitest tests would hang in watch mode waiting for user input. Tests now run to completion automatically without requiring interaction.
245+
**Contributors**: KJ7LNW
246+
247+
---
248+
249+
## PR #5279: fix: move context condensing prompt to Prompts section (#4924)
250+
251+
**Author**: SannidhyaSah
252+
**Linked Issues**: #4924 (reporter: notadamking)
253+
**Category**: QOL Improvement
254+
**User Impact**: The context condensing prompt setting is now easier to find in the Prompts section where users naturally look for it, and the prompt remains visible regardless of auto-condensing settings
255+
**Contributors**: SannidhyaSah, notadamking
256+
257+
---
258+
259+
## PR #5946: feat: Add Mistral embedding provider (#5932)
260+
261+
**Author**: SannidhyaSah
262+
**Linked Issues**: #5932 (reporter: shariqriazz)
263+
**Category**: Feature
264+
**User Impact**: Adds Mistral as a new embedding provider for codebase indexing, offering superior performance at no cost. Users can now choose Mistral's codestral-embed model for better code understanding and more accurate AI responses while reducing operational costs.
265+
**Contributors**: SannidhyaSah, shariqriazz
266+
267+
---

‎docs/update-notes/v3.23.17.mdx‎

Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
---
2+
description: This release introduces intelligent merge conflict resolution, new AI providers, enhanced stability, and numerous quality-of-life improvements.
3+
keywords:
4+
- roo code 3.23.17
5+
- merge resolver
6+
- mistral embedding
7+
- moonshot provider
8+
- qwen model
9+
image: /img/social-share.jpg
10+
---
11+
12+
# Roo Code 3.23.17 Release Notes (2025-07-22)
13+
14+
This release introduces new AI providers, enhanced stability, and numerous quality-of-life improvements.
15+
16+
## QOL Improvements
17+
18+
* **Jump to New Files**: Added jump icon for newly created files, matching the experience of edited files (thanks mkdir700!) ([#5738](https://github.com/RooCodeInc/Roo-Code/pull/5738))
19+
* **Apply Diff Error Messages**: Added case sensitivity reminder when apply_diff fails, helping users understand matching requirements (thanks maskelihileci!) ([#6076](https://github.com/RooCodeInc/Roo-Code/pull/6076))
20+
* **Context Condensing Prompt Location**: Moved to Prompts section for better discoverability and persistent visibility (thanks SannidhyaSah, notadamking!) ([#5279](https://github.com/RooCodeInc/Roo-Code/pull/5279))
21+
* **Todo List Tool Control**: Added checkbox in provider settings to enable/disable the todo list tool ([#6032](https://github.com/RooCodeInc/Roo-Code/pull/6032))
22+
* **MCP Content Optimization**: Automatically omits MCP-related prompts when no servers are configured ([#5889](https://github.com/RooCodeInc/Roo-Code/pull/5889))
23+
* **Git Installation Check**: Shows clear warning with download link when Git is not installed for checkpoints feature (thanks MuriloFP!) ([#5920](https://github.com/RooCodeInc/Roo-Code/pull/5920))
24+
* **Configurable Eval Timeouts**: Added slider to set evaluation timeouts between 5-10 minutes ([#5865](https://github.com/RooCodeInc/Roo-Code/pull/5865))
25+
26+
### Bug Fixes
27+
28+
* **Terminal Output Limits**: Fixed critical crashes caused by terminal commands outputting extremely long lines. Now properly limits output to prevent context explosion (thanks robottwo!) ([#5777](https://github.com/RooCodeInc/Roo-Code/pull/5777))
29+
* **API Error Messages**: Fixed issue where all API failures showed as "API Request Cancelled". Now displays correct error messages for better troubleshooting (thanks MDean-Slalom, daniel-lxs!) ([#6025](https://github.com/RooCodeInc/Roo-Code/pull/6025))
30+
* **AWS Bedrock Token Counting**: Fixed inaccurate token counts and false context window errors for Bedrock users (thanks daniel-lxs, fedorbass!) ([#6019](https://github.com/RooCodeInc/Roo-Code/pull/6019))
31+
* **Export During API Operations**: Restored ability to export, share, and copy content while API is running (thanks MuriloFP, KJ7LNW!) ([#5849](https://github.com/RooCodeInc/Roo-Code/pull/5849))
32+
* **Symlinked Rules Files**: Fixed sorting of symlinked rules files to use symlink names, allowing proper ordering with numbered prefixes (thanks dsent!) ([#5903](https://github.com/RooCodeInc/Roo-Code/pull/5903))
33+
* **Global Mode Export**: Fixed issue where exporting global modes didn't include associated rules files ([#5837](https://github.com/RooCodeInc/Roo-Code/pull/5837))
34+
* **Vitest Command**: Added missing run parameter to prevent tests hanging in watch mode (thanks KJ7LNW!) ([#5991](https://github.com/RooCodeInc/Roo-Code/pull/5991))
35+
* **Evaluation System**: Fixed evaluation timeout handling that was broken in a previous update ([#6065](https://github.com/RooCodeInc/Roo-Code/pull/6065))
36+
* **Cloud Session Handling**: Improved session error handling to prevent unnecessary logouts during service issues ([#6051](https://github.com/RooCodeInc/Roo-Code/pull/6051))
37+
38+
## Provider Updates
39+
40+
* **Mistral Embedding Provider**: Codebase indexing gets a major upgrade with Mistral as a new embedding provider, offering superior performance at no cost. Simply select Mistral's codestral-embed model in your embedding settings for better code understanding and more accurate AI responses (thanks SannidhyaSah, shariqriazz!) ([#5946](https://github.com/RooCodeInc/Roo-Code/pull/5946))
41+
* **Qwen3-235B Model**: Added support for Qwen3-235B-A22B-Instruct-2507 with massive 262K token context window on Chutes AI (thanks apple-techie!) ([#6052](https://github.com/RooCodeInc/Roo-Code/pull/6052))
42+
* **Moonshot AI**: Added Moonshot as a new AI provider option (thanks CellenLee!) ([#6046](https://github.com/RooCodeInc/Roo-Code/pull/6046))
43+
* **Llama 4 Maverick**: Added Llama 4 Maverick model (128K context) to Google Vertex AI in US-EAST5 region (thanks MuriloFP, konstantinosbotonakis!) ([#6023](https://github.com/RooCodeInc/Roo-Code/pull/6023))
44+
45+
## Misc Improvements
46+
47+
* **Merge Resolver Mode**: Added intelligent merge conflict resolution mode that analyzes commit history to make smart decisions about which changes to keep (thanks daniel-lxs!) ([#6090](https://github.com/RooCodeInc/Roo-Code/pull/6090))
48+
* **Token Fallback Logic**: Improved sliding window with consistent 8192 token fallback for more predictable context handling ([#5993](https://github.com/RooCodeInc/Roo-Code/pull/5993))
49+
* **Todo List Clarification**: Updated documentation to clarify that todo list tool is for complicated tasks, not just multi-step ones ([#5926](https://github.com/RooCodeInc/Roo-Code/pull/5926))

0 commit comments

Comments
 (0)