Add release notes for v3.28.8 (#364)

- Created docs/update-notes/v3.28.8.mdx with expanded 'No tool used' error fixes
- Updated docs/update-notes/index.md with v3.28.8 entry
- Updated sidebars.ts to include v3.28.8 in navigation
- Updated combined v3.28.mdx with all v3.28.8 changes
- Added hero image to static/img/v3.28.8/v3.28.8.png

Author: hannesrudolph

docs/update-notes/index.md

@@ -20,6 +20,7 @@ image: /img/social-share.jpg
 ### Version 3.28
 
 *   [3.28](/update-notes/v3.28) (Combined)
+*   [3.28.8](/update-notes/v3.28.8) (2025-09-25)
 *   [3.28.7](/update-notes/v3.28.7) (2025-09-23)
 *   [3.28.6](/update-notes/v3.28.6) (2025-09-23)
 *   [3.28.5](/update-notes/v3.28.5) (2025-09-20)

docs/update-notes/v3.28.8.mdx

@@ -0,0 +1,41 @@
+---
+description: Bug fixes for tool usage errors and context condensing, plus new free models and cloud account management improvements.
+keywords:
+  - roo code 3.28.8
+  - tool usage fixes
+  - context condensing
+  - free models
+  - cloud accounts
+image: /img/v3.28.8/v3.28.8.png
+---
+
+# Roo Code 3.28.8 Release Notes (2025-09-25)
+
+This release fixes critical GPT-5-Codex tool usage errors, improves context condensing, and adds new free models to the Roo provider.
+
+## Fixed: GPT-5-Codex errors and more!
+
+We've resolved the situation where LLMs would sometimes not make tool calls in their repsonse which improves Roo's overall flow. ([#8292](https://github.com/RooCodeInc/Roo-Code/pull/8292))
+
+- **GPT-5-Codex Response Fix**: Fixes the issue where the model `gpt-5-codex` (and less frequently other models) would send a response with no assistant message
+- **Completion Tool Reliability**: This fix eliminates some situations where the LLM would repeat its last text message when the attempt_completion tool was called
+- **Tool Call Consistency**: Fixes where the LLM would sometimes respond without a tool call
+
+## QOL Improvements
+
+* **Cloud Account Management**: Added account switcher and "Create Team Account" option for logged-in users, making it easier to manage multiple accounts ([#8291](https://github.com/RooCodeInc/Roo-Code/pull/8291))
+* **Social Media Sharing**: Improved OpenGraph images for better preview cards when sharing Roo Code links on Slack, Twitter, and Facebook ([#8285](https://github.com/RooCodeInc/Roo-Code/pull/8285))
+
+## Bug Fixes
+
+* **Context Condensing**: Fixed an issue where the initial task request was being lost during context condensing, causing Roo to try to re-answer the original task ask when resuming after condensing([#8298](https://github.com/RooCodeInc/Roo-Code/pull/8298))
+
+## Provider Updates
+
+* **New Free Models**: Added two more free models to the Roo provider ([#8304](https://github.com/RooCodeInc/Roo-Code/pull/8304)):
+  - xai/grok-4-fast
+  - deepseek/deepseek-chat-v3.1
+
+## Misc Improvements
+
+* **Infrastructure**: Set port 3446 for web-evals service in production mode ([#8288](https://github.com/RooCodeInc/Roo-Code/pull/8288))

docs/update-notes/v3.28.mdx

@@ -64,6 +64,8 @@ This gives teams a higher-capacity OpenAI option without extra configuration.[#8
 
 ## QOL Improvements
 
+* **Cloud Account Management**: Added account switcher and "Create Team Account" option for logged-in users, making it easier to manage multiple accounts ([#8291](https://github.com/RooCodeInc/Roo-Code/pull/8291))
+* **Social Media Sharing**: Improved OpenGraph images for better preview cards when sharing Roo Code links on Slack, Twitter, and Facebook ([#8285](https://github.com/RooCodeInc/Roo-Code/pull/8285))
 * **Auto-approve keyboard shortcut**: Toggle approvals with Cmd/Ctrl+Alt+A from anywhere in the editor so you can stay in the flow while reviewing changes (via [#8214](https://github.com/RooCodeInc/Roo-Code/pull/8214))
 * **Click-to-Edit Chat Messages**: Click directly on any message text to edit it, with ESC to cancel and improved padding consistency ([#7790](https://github.com/RooCodeInc/Roo-Code/pull/7790))
 * **Enhanced Reasoning Display**: The AI's thinking process now shows a persistent timer and displays reasoning content in clean italic text ([#7752](https://github.com/RooCodeInc/Roo-Code/pull/7752))
@@ -91,6 +93,8 @@ This gives teams a higher-capacity OpenAI option without extra configuration.[#8
 
 ## Bug Fixes
 
+* **Fixed "No tool used" Errors**: Resolved the situation where LLMs would sometimes not make tool calls in their repsonse which improves Roo's overall flow. ([#8292](https://github.com/RooCodeInc/Roo-Code/pull/8292))
+* **Context Condensing**: Fixed an issue where the initial task request was being lost during context condensing, causing Roo to try to re-answer the original task ask when resuming after condensing([#8298](https://github.com/RooCodeInc/Roo-Code/pull/8298))
 * **Roo provider stays signed in**: Roo provider tokens refresh automatically and the local evals app binds to port 3446 for predictable scripts (via [#8224](https://github.com/RooCodeInc/Roo-Code/pull/8224))
 * **Checkpoint text stays on one line**: Prevented multi-line wrapping in languages such as Chinese, Korean, Japanese, and Russian so the checkpoint UI stays compact (via [#8207](https://github.com/RooCodeInc/Roo-Code/pull/8207); reported in [#8206](https://github.com/RooCodeInc/Roo-Code/issues/8206))
 * **Ollama respects Modelfile num_ctx**: Roo now defers to your Modelfile’s context window to avoid GPU OOMs while still allowing explicit overrides when needed (via [#7798](https://github.com/RooCodeInc/Roo-Code/pull/7798); reported in [#7797](https://github.com/RooCodeInc/Roo-Code/issues/7797))
@@ -120,6 +124,7 @@ This gives teams a higher-capacity OpenAI option without extra configuration.[#8
 
 ## Provider Updates
 
+* **New Free Models**: Added two more free models to the Roo provider - xai/grok-4-fast and deepseek/deepseek-chat-v3.1 ([#8304](https://github.com/RooCodeInc/Roo-Code/pull/8304))
 * **Vertex AI Models**: Added support for 6 new models including DeepSeek-V3, GPT-OSS, and Qwen models, plus the us-south1 region (thanks ssweens!) ([#7727](https://github.com/RooCodeInc/Roo-Code/pull/7727))
 * **DeepSeek Pricing**: Updated to new unified rates effective September 5, 2025 - $0.56/M input tokens, $1.68/M output tokens (thanks NaccOll!) ([#7687](https://github.com/RooCodeInc/Roo-Code/pull/7687))
 
@@ -129,6 +134,7 @@ This gives teams a higher-capacity OpenAI option without extra configuration.[#8
 
 ## Misc Improvements
 
+* **Infrastructure**: Set port 3446 for web-evals service in production mode ([#8288](https://github.com/RooCodeInc/Roo-Code/pull/8288))
 * **Cloud Task Button**: Added a new button for opening tasks in Roo Code Cloud with QR codes and shareable URLs ([#7572](https://github.com/RooCodeInc/Roo-Code/pull/7572))
 * **Telemetry Default**: Posthog telemetry is now enabled by default with easy opt-out options in settings ([#7909](https://github.com/RooCodeInc/Roo-Code/pull/7909))
 * **Roo Code Cloud Announcement**: Updated in-app announcements about Roo Code Cloud features, localized in 18 languages ([#7914](https://github.com/RooCodeInc/Roo-Code/pull/7914))

sidebars.ts

@@ -223,6 +223,7 @@ const sidebars: SidebarsConfig = {
           label: '3.28',
           items: [
             { type: 'doc', id: 'update-notes/v3.28', label: '3.28 Combined' },
+            { type: 'doc', id: 'update-notes/v3.28.8', label: '3.28.8' },
             { type: 'doc', id: 'update-notes/v3.28.7', label: '3.28.7' },
             { type: 'doc', id: 'update-notes/v3.28.6', label: '3.28.6' },
             { type: 'doc', id: 'update-notes/v3.28.5', label: '3.28.5' },