Merge remote-tracking branch 'upstream/main'

Author: Roo Agent

.env.sample

@@ -3,3 +3,4 @@ POSTHOG_API_KEY=key-goes-here
 # Roo Code Cloud / Local Development
 CLERK_BASE_URL=https://epic-chamois-85.clerk.accounts.dev
 ROO_CODE_API_URL=http://localhost:3000
+ROO_CODE_PROVIDER_URL=http://localhost:8080/proxy/v1

.gitattributes

@@ -6,6 +6,11 @@ src/assets/docs/demo.gif filter=lfs diff=lfs merge=lfs -text
 *.snap linguist-generated=true
 
 # Non-English translation files - mark as linguist-generated to exclude from GitHub language statistics
+# Package NLS files - mark non-English ones as generated
+src/package.nls.*.json linguist-generated=true
+# Exclude the base English file from being marked as generated
+src/package.nls.json linguist-generated=false
+
 # Root locales directory (contains only non-English translations)
 locales/** linguist-generated=true
 

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -25,6 +25,7 @@ body:
         - AWS Bedrock
         - Chutes AI
         - DeepSeek
+        - Featherless AI
         - Fireworks AI
         - Glama
         - Google Gemini

.roo/commands/release.md

@@ -4,7 +4,7 @@ argument-hint: patch | minor | major
 ---
 
 1. Identify the SHA corresponding to the most recent release using GitHub CLI: `gh release view --json tagName,targetCommitish,publishedAt`
-2. Analyze changes since the last release using: `gh pr list --state merged --json number,title,author,url,mergedAt,closingIssuesReferences --limit 1000 -q '[.[] | select(.mergedAt > "TIMESTAMP") | {number, title, author: .author.login, url, mergedAt, issues: .closingIssuesReferences}] | sort_by(.number)'`
+2. Analyze changes since the last release using: `gh pr list --state merged --base main --json number,title,author,url,mergedAt,closingIssuesReferences --limit 1000 -q '[.[] | select(.mergedAt > "TIMESTAMP") | {number, title, author: .author.login, url, mergedAt, issues: .closingIssuesReferences}] | sort_by(.number)'`
 3. For each PR with linked issues, fetch the issue details to get the issue reporter: `gh issue view ISSUE_NUMBER --json number,author -q '{number, reporter: .author.login}'`
 4. Summarize the changes. If the user did not specify, ask them whether this should be a major, minor, or patch release.
 5. Create a changeset in .changeset/v[version].md instead of directly modifying package.json. The format is:

.roo/roomotes.yml

@@ -20,14 +20,6 @@ github_events:
     - event: pull_request.opened
       action:
           name: github.pr.review
-    - event: pull_request.opened
-      action:
-          name: general.task
-          prompt: |
-              1. Run the script `node scripts/find-missing-translations.js` and carefully review its output for any missing translations.
-              2. If the script reports missing translations, switch into `translate` mode and add them in all supported languages.
-              3. If you've added new translations, commit and push them to the existing PR.
-              4. If you get a permission error trying to push to the PR just give up (i.e don't create a new PR instead).
     - event: pull_request_review_comment.created
       action:
           name: github.pr.comment.respond

.roo/rules-issue-fixer/1_Workflow.xml

@@ -273,6 +273,13 @@
        - [Note any technical terms or constraints]
 
        Please ensure all translations maintain consistency with existing terminology and follow the project's localization guidelines.</message>
+       <todos>
+[ ] Identify all user-facing strings that need translation
+[ ] Update i18n JSON files for all supported languages
+[ ] Translate any markdown documentation files
+[ ] Verify translations maintain consistency with existing terminology
+[ ] Test translations in the application context
+       </todos>
        </new_task>
 
        Wait for the translation task to complete before proceeding to testing.

.roo/rules-issue-fixer/2_best_practices.xml

@@ -14,5 +14,6 @@
   - Check for accessibility issues (for UI changes)
   - Delegate translation tasks to translate mode when implementing user-facing changes
   - Always check for hard-coded strings and internationalization needs
+  - When using new_task to delegate work, always include a comprehensive todos list
   - Wait for translation completion before proceeding to final testing
 </best_practices>

CHANGELOG.md

@@ -1,5 +1,114 @@
 # Roo Code Changelog
 
+## [3.25.23] - 2025-08-22
+
+- feat: add custom base URL support for Requesty provider (thanks @requesty-JohnCosta27!)
+- feat: add DeepSeek V3.1 model to Chutes AI provider (#7294 by @dmarkey, PR by @app/roomote)
+- Revert "feat: enable loading Roo modes from multiple files in .roo/modes directory" temporarily to fix a bug with mode installation
+
+## [3.25.22] - 2025-08-22
+
+- Add prompt caching support for Kimi K2 on Groq (thanks @daniel-lxs and @benank!)
+- Add documentation links for global custom instructions in UI (thanks @app/roomote!)
+
+## [3.25.21] - 2025-08-21
+
+- Ensure subtask results are provided to GPT-5 in OpenAI Responses API
+- Promote the experimental AssistantMessageParser to the default parser
+- Update DeepSeek models context window to 128k (thanks @JuanPerezReal)
+- Enable grounding features for Vertex AI (thanks @anguslees)
+- Allow orchestrator to pass TODO lists to subtasks
+- Improved MDM handling
+- Handle nullish token values in ContextCondenseRow to prevent UI crash (thanks @s97712)
+- Improved context window error handling for OpenAI and other providers
+- Add "installed" filter to Roo Marketplace (thanks @semidark)
+- Improve filesystem access checks (thanks @elianiva)
+- Support for loading Roo modes from multiple YAML files in the `.roo/modes/` directory (thanks @farazoman)
+- Add Featherless provider (thanks @DarinVerheijke)
+
+## [3.25.20] - 2025-08-19
+
+- Add announcement for Sonic model
+
+## [3.25.19] - 2025-08-19
+
+- Fix issue where new users couldn't select the Roo Code Cloud provider (thanks @daniel-lxs!)
+
+## [3.25.18] - 2025-08-19
+
+- Add new stealth Sonic model through the Roo Code Cloud provider
+- Fix: respect enableReasoningEffort setting when determining reasoning usage (#7048 by @ikbencasdoei, PR by @app/roomote)
+- Fix: prevent duplicate LM Studio models with case-insensitive deduplication (#6954 by @fbuechler, PR by @daniel-lxs)
+- Feat: simplify ask_followup_question prompt documentation (thanks @daniel-lxs!)
+- Feat: simple read_file tool for single-file-only models (thanks @daniel-lxs!)
+- Fix: Add missing zaiApiKey and doubaoApiKey to SECRET_STATE_KEYS (#7082 by @app/roomote)
+- Feat: Add new models and update configurations for vscode-lm (thanks @NaccOll!)
+
+## [3.25.17] - 2025-08-17
+
+- Fix: Resolve terminal reuse logic issues
+
+## [3.25.16] - 2025-08-16
+
+- Add support for OpenAI gpt-5-chat-latest model (#7057 by @PeterDaveHello, PR by @app/roomote)
+- Fix: Use native Ollama API instead of OpenAI compatibility layer (#7070 by @LivioGama, PR by @daniel-lxs)
+- Fix: Prevent XML entity decoding in diff tools (#7107 by @indiesewell, PR by @app/roomote)
+- Fix: Add type check before calling .match() on diffItem.content (#6905 by @pwilkin, PR by @app/roomote)
+- Refactor task execution system: improve call stack management (thanks @catrielmuller!)
+- Fix: Enable save button for provider dropdown and checkbox changes (thanks @daniel-lxs!)
+- Add an API for resuming tasks by ID (thanks @mrubens!)
+- Emit event when a task ask requires interaction (thanks @cte!)
+- Make enhance with task history default to true (thanks @liwilliam2021!)
+- Fix: Use cline.cwd as primary source for workspace path in codebaseSearchTool (thanks @NaccOll!)
+- Hotfix multiple folder workspace checkpoint (thanks @NaccOll!)
+
+## [3.25.15] - 2025-08-14
+
+- Fix: Remove 500-message limit to prevent scrollbar jumping in long conversations (#7052, #7063 by @daniel-lxs, PR by @app/roomote)
+- Fix: Reset condensing state when switching tasks (#6919 by @f14XuanLv, PR by @f14XuanLv)
+- Fix: Implement sitemap generation in TypeScript and remove XML file (#5231 by @abumalick, PR by @abumalick)
+- Fix: allowedMaxRequests and allowedMaxCost values not showing in the settings UI (thanks @chrarnoldus!)
+
+## [3.25.14] - 2025-08-13
+
+- Fix: Only include verbosity parameter for models that support it (#7054 by @eastonmeth, PR by @app/roomote)
+- Fix: AWS Bedrock 1M context - Move anthropic_beta to additionalModelRequestFields (thanks @daniel-lxs!)
+- Fix: Make cancelling requests more responsive by reverting recent changes
+
+## [3.25.13] - 2025-08-12
+
+- Add Sonnet 1M context checkbox to Bedrock
+- Fix: add --no-messages flag to ripgrep to suppress file access errors (#6756 by @R-omk, PR by @app/roomote)
+- Add support for AGENT.md alongside AGENTS.md (#6912 by @Brendan-Z, PR by @app/roomote)
+- Remove deprecated GPT-4.5 Preview model (thanks @PeterDaveHello!)
+
+## [3.25.12] - 2025-08-12
+
+- Update: Claude Sonnet 4 context window configurable to 1 million tokens in Anthropic provider (thanks @daniel-lxs!)
+- Add: Minimal reasoning support to OpenRouter (thanks @daniel-lxs!)
+- Fix: Add configurable API request timeout for local providers (#6521 by @dabockster, PR by @app/roomote)
+- Fix: Add --no-sandbox flag to browser launch options (#6632 by @QuinsZouls, PR by @QuinsZouls)
+- Fix: Ensure JSON files respect .rooignore during indexing (#6690 by @evermoving, PR by @app/roomote)
+- Add: New Chutes provider models (#6698 by @fstandhartinger, PR by @app/roomote)
+- Add: OpenAI gpt-oss models to Amazon Bedrock dropdown (#6752 by @josh-clanton-powerschool, PR by @app/roomote)
+- Fix: Correct tool repetition detector to not block first tool call when limit is 1 (#6834 by @NaccOll, PR by @app/roomote)
+- Fix: Improve checkpoint service initialization handling (thanks @NaccOll!)
+- Update: Improve zh-TW Traditional Chinese locale (thanks @PeterDaveHello!)
+- Add: Task expand and collapse translations (thanks @app/roomote!)
+- Update: Exclude GPT-5 models from 20% context window output token cap (thanks @app/roomote!)
+- Fix: Truncate long model names in model selector to prevent overflow (thanks @app/roomote!)
+- Add: Requesty base url support (thanks @requesty-JohnCosta27!)
+
+## [3.25.11] - 2025-08-11
+
+- Add: Native OpenAI provider support for Codex Mini model (#5386 by @KJ7LNW, PR by @daniel-lxs)
+- Add: IO Intelligence Provider support (thanks @ertan2002!)
+- Fix: MCP startup issues and remove refresh notifications (thanks @hannesrudolph!)
+- Fix: Improvements to GPT-5 OpenAI provider configuration (thanks @hannesrudolph!)
+- Fix: Clarify codebase_search path parameter as optional and improve tool descriptions (thanks @app/roomote!)
+- Fix: Bedrock provider workaround for LiteLLM passthrough issues (thanks @jr!)
+- Fix: Token usage and cost being underreported on cancelled requests (thanks @chrarnoldus!)
+
 ## [3.25.10] - 2025-08-07
 
 - Add support for GPT-5 (thanks Cline and @app/roomote!)

PRIVACY.md

@@ -1,14 +1,14 @@
 # Roo Code Privacy Policy
 
-**Last Updated: June 10th, 2025**
+**Last Updated: August 20th, 2025**
 
 Roo Code respects your privacy and is committed to transparency about how we handle your data. Below is a simple breakdown of where key pieces of data go—and, importantly, where they don’t.
 
 ### **Where Your Data Goes (And Where It Doesn’t)**
 
-- **Code & Files**: Roo Code accesses files on your local machine when needed for AI-assisted features. When you send commands to Roo Code, relevant files may be transmitted to your chosen AI model provider (e.g., OpenAI, Anthropic, OpenRouter) to generate responses. We do not have access to this data, but AI providers may store it per their privacy policies.
+- **Code & Files**: Roo Code accesses files on your local machine when needed for AI-assisted features. When you send commands to Roo Code, relevant files may be transmitted to your chosen AI model provider (e.g., OpenAI, Anthropic, OpenRouter) to generate responses. If you select Roo Code Cloud as the model provider (proxy mode), your code may transit Roo Code servers only to forward it to the upstream provider. We do not store your code; it is deleted immediately after forwarding. Otherwise, your code is sent directly to the provider. AI providers may store data per their privacy policies.
 - **Commands**: Any commands executed through Roo Code happen on your local environment. However, when you use AI-powered features, the relevant code and context from your commands may be transmitted to your chosen AI model provider (e.g., OpenAI, Anthropic, OpenRouter) to generate responses. We do not have access to or store this data, but AI providers may process it per their privacy policies.
-- **Prompts & AI Requests**: When you use AI-powered features, your prompts and relevant project context are sent to your chosen AI model provider (e.g., OpenAI, Anthropic, OpenRouter) to generate responses. We do not store or process this data. These AI providers have their own privacy policies and may store data per their terms of service.
+- **Prompts & AI Requests**: When you use AI-powered features, your prompts and relevant project context are sent to your chosen AI model provider (e.g., OpenAI, Anthropic, OpenRouter) to generate responses. We do not store or process this data. These AI providers have their own privacy policies and may store data per their terms of service. If you choose Roo Code Cloud as the provider (proxy mode), prompts may transit Roo Code servers only to forward them to the upstream model and are not stored.
 - **API Keys & Credentials**: If you enter an API key (e.g., to connect an AI model), it is stored locally on your device and never sent to us or any third party, except the provider you have chosen.
 - **Telemetry (Usage Data)**: We only collect feature usage and error data if you explicitly opt-in. This telemetry is powered by PostHog and helps us understand feature usage to improve Roo Code. This includes your VS Code machine ID and feature usage patterns and exception reports. We do **not** collect personally identifiable information, your code, or AI prompts.
 - **Marketplace Requests**: When you browse or search the Marketplace for Model Configuration Profiles (MCPs) or Custom Modes, Roo Code makes a secure API call to Roo Code’s backend servers to retrieve listing information. These requests send only the query parameters (e.g., extension version, search term) necessary to fulfill the request and do not include your code, prompts, or personally identifiable information.