feat: merge roocode (zgsm-ai#155)

* Rename cline_docs -> docs (RooCodeInc#3587)

* Update contributors list (RooCodeInc#3299)

Co-authored-by: mrubens <[email protected]>

* fix(deps): update dependency posthog-js to v1.242.1 (RooCodeInc#3602)

Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

* Use a shadcn dialog for the announcement (RooCodeInc#3604)

* feat: add buildDocLink utility and 21 Internal Links to Docs (RooCodeInc#3418)

Co-authored-by: Matt Rubens <[email protected]>

* Add build vsix Workflow (RooCodeInc#3600)

* build: enable source maps for improved debugging (RooCodeInc#3596)

Co-authored-by: Eric Wheeler <[email protected]>

* v3.16.7 (RooCodeInc#3614)

* [Condense] Condense messages with an LLM rather than truncating (RooCodeInc#3582)

Co-authored-by: Matt Rubens <[email protected]>

* Fix type generation (RooCodeInc#3619)

* Update contributors list (RooCodeInc#3612)

Co-authored-by: mrubens <[email protected]>

* v3.17.0 (RooCodeInc#3622)

* Changeset version bump (RooCodeInc#3556)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Matt Rubens <[email protected]>

* fix: correct Changelog link in localized README files (RooCodeInc#3629)

The Changelog link in `locales/ja/README.md` and other localized
READMEswas pointing to a broken relative path, resulting in 404s.This
commit updates the link to use a correct relative path
(`../../CHANGELOG.md`)so that it works across all locales.

* Fix incorrect reserved tokens calculation on OpenRouter (RooCodeInc#3626)

fix: improve token reservation logic in calculateTokenDistribution

* Fix command display in the approval required case (RooCodeInc#3636)

* Changeset version bump (RooCodeInc#3637)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: R00-B0T <[email protected]>
Co-authored-by: Chris Estreich <[email protected]>

* Fix how custom instructions are loaded into the API request (RooCodeInc#3638)p

* Lock the versions of vsce and ovsx (RooCodeInc#3643)

* Revert "Switch to the new Roo message parser" (RooCodeInc#3649)

* Changeset version bump (RooCodeInc#3645)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Chris Estreich <[email protected]>

* Import settings bug fix / improvements (RooCodeInc#3657)

* Export ProviderName type to Roo-Code-Types (RooCodeInc#3675)

* Log Cleanup to Remove Cline (RooCodeInc#3704)

* Rename Errors & Fix Spelling Mistake

* Update src/core/task/Task.ts

Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>

---------

Co-authored-by: Matt Rubens <[email protected]>
Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>

* RooCodeInc#3679 - Fixes packaging to include correct tiktoken.wasm (lite) (RooCodeInc#3697)

- also, additions to .gitignore and .vscodeignore to prevent the IntelliJ .idea and .qodo folders from being included for git and packaging.

* Adds refresh models button for Unbound provider (RooCodeInc#3663)

* Adds refresh models button for Unbound provider

* Adds changeset

* Optimizes code to prevent memory leak, add error messages

* Adds unbound messages to all supported languages

---------

Co-authored-by: Pugazhendhi <[email protected]>

* Add Qwen3 model series to the Chutes provider (RooCodeInc#3710)

* Add Qwen3 model series to the Chutes provider

New models for the Chutes provider:

- Qwen/Qwen3-235B-A22B
- Qwen/Qwen3-32B
- Qwen/Qwen3-30B-A3B
- Qwen/Qwen3-14B
- Qwen/Qwen3-8B

* add changeset

* fix(webview): Fix links to filename:0 (RooCodeInc#3727)

* fix(webview): Fix links to filename:0

* Add changeset

* LM studio reasoning support (thinking block) (RooCodeInc#3719)

lmstudio reasoning support (thinking block)

Similar to ollama implementation in RooCodeInc#1080

* feat(evals): add UI and backend support for importing and injecting f… (RooCodeInc#3606)

* [Condense Context] Track metrics around context condensing and show in UI

* Add UI component

* account for system prompt when estimating new context size

* add header

* bug fix

* nit

* nit

* refactor

* fix

* add unit tests for condense

* update sliding-window tests

* add getApiMetrics.test.ts

* fix failing tests

* use chat.json

* add translations

* add tests for ContextCondenseRow

* add changeset

* camelCase

* use Markdown for summary

* use tailwind

* non default export

* rm test :/

* Make prompt input textareas resizable (RooCodeInc#3691) (RooCodeInc#3739)

* feat: move play audio to webview to ensure cross-platform (RooCodeInc#3659)

Co-authored-by: sam hoang <[email protected]>

* refactor:  import multiple times (RooCodeInc#3745)

* Add YAML support for .roomode files alongside JSON processing (RooCodeInc#3711)

* ✨ feat(settings): Add allowedMaxRequests feature inspired by Cline (RooCodeInc#3631)

* feat(settings): Introduce the "auto-approve request count" feature from Cline

This is the first minor UI feature I've added, so please let me know if I'm missing anything! (translations, organization, etc!)

Please see commits for details

introduce allowedMaxRequests to globalSettingsSchema
update ExtensionState and its context with allowedMaxRequests
implement UI for setting max requests in AutoApproveMenu component
prompt user when auto-approval limit is reached with i18n support
increment consecutiveAutoApprovedRequestsCount and reset upon user approval
add translations for auto-approved request limit reached prompt in multiple languages
add new UI for "auto_approval_max_req_reached" in ChatRowContent
display prompt with title, description, and button for user action

🔧 chore(gitignore): add .idea to .gitignore to exclude IDE-specific files
- remove .idea/workspace.xml to clean up repository

* 🔧 chore(gitignore): add IDE configuration files to ignore list

- add .idea directory to ignore JetBrains IDE configurations

* 🌐 i18n(chat): add translation keys for api request limit

- introduce translation keys for "title" and "unlimited" in multiple languages
- update description for api request limit in various languages

* 🌐 i18n(chat): migrate auto-approved request limit translations

- move translations from common.json to chat.json across locales
- update component to use Trans for dynamic text rendering

* Update the UI for setting max requests

* Hide the auto-approve limit warning once clicked

---------

Co-authored-by: Matt Rubens <[email protected]>

* Move error message for settings import failure into the correct position (RooCodeInc#3752)

Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>
Co-authored-by: Chris Estreich <[email protected]>

* feat: use template variables for version numbers in announcement strings (RooCodeInc#3755)

* Auto-reload core changes in dev mode (RooCodeInc#3284)

Co-authored-by: Matt Rubens <[email protected]>

* Moved repo to new org (RooCodeInc#3756)

* Use yaml as default custom modes format (RooCodeInc#3749)

* [Condense] Add a button to condense the task context (RooCodeInc#3623)

* [Condense] Add a button to condense the task context

* wip

* wip

* wip

* bring back delete size

* account for the system prompt in the context

* update tests to use systemPrompt

* add type

* translations

* nit

* update tests

* filter to the current task

* nit

* refactor

* nit

* non interactive option

* simplify chat summary UI

* changeset

* nit

* fix check-types

* throw

* [Condense] Fix double counting last message when condensing (RooCodeInc#3763)

* Get package publisher and name from package.json + command type safety (RooCodeInc#3766)

* Lm studio and ollama usage fix (RooCodeInc#3707)

* integration

* Fix

* [Condense] Change condense icon (RooCodeInc#3768)

* [Condense] Change condense icon

* change to fold

* feat: add gemini-2.5-flash-preview-05-20 models (RooCodeInc#3769)

* Add Gemini Flash 2.5 05-20 variants for the Vertex provider (RooCodeInc#3758)

* feat(api): add gemini-2.5-flash-preview-05-20 model configuration

* feat(tests): update apiModelId to gemini-2.5-flash-preview-05-20 in ProviderSettingsManager tests in case the old version is deprecated

* chore: add changeset

* feat(api): update vertexModels to add gemini-2.5-flash-preview-05-20 variants

* chore: update changeset

* [Condense] Show indicator message when context is condensing (RooCodeInc#3765)

* [Condense] Show indicator message when context is condensing

* changeset

* translations

* Another grey screen fix. (RooCodeInc#3644)

Memory memory memory

* Fix: Missing or inconsistent syntax highlighting across UI components (RooCodeInc#3656)

* fix: Missing or inconsistent syntax highlighting across UI components

- Change file listings to use 'shellsession' for terminal-like highlighting
- Use 'markdown' for code definitions and instructions
- Add file extension-based language detection for new files
- Ensure consistent 'diff' highlighting for all diff content
- Use 'xml' language for error messages
- Make language property required in CodeAccordian
- Set default fallback to 'txt' instead of undefined

Fixes: RooCodeInc#3655
Signed-off-by: Eric Wheeler <[email protected]>

* chore: make language property required in CodeBlock

- Updated CodeBlockProps interface to make language property required
- Updated mock implementation to match the interface change
- Ensured CodeAccordian always provides a fallback language value

Signed-off-by: Eric Wheeler <[email protected]>

---------

Signed-off-by: Eric Wheeler <[email protected]>
Co-authored-by: Eric Wheeler <[email protected]>

* Add contact section to pull request template for communication (RooCodeInc#3771)

* Update contributors list (RooCodeInc#3620)

Co-authored-by: mrubens <[email protected]>

* More VSCode command / build fixes (RooCodeInc#3780)

---------

Signed-off-by: Eric Wheeler <[email protected]>
Co-authored-by: Matt Rubens <[email protected]>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
Co-authored-by: Chris Estreich <[email protected]>
Co-authored-by: Hannes Rudolph <[email protected]>
Co-authored-by: மனோஜ்குமார் பழனிச்சாமி <[email protected]>
Co-authored-by: KJ7LNW <[email protected]>
Co-authored-by: Eric Wheeler <[email protected]>
Co-authored-by: Canyon Robins <[email protected]>
Co-authored-by: R00-B0T <[email protected]>
Co-authored-by: hatsu <[email protected]>
Co-authored-by: Daniel <[email protected]>
Co-authored-by: R00-B0T <[email protected]>
Co-authored-by: xyOz <[email protected]>
Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>
Co-authored-by: vagadiya <[email protected]>
Co-authored-by: pugazhendhi-m <[email protected]>
Co-authored-by: Pugazhendhi <[email protected]>
Co-authored-by: zeo <[email protected]>
Co-authored-by: Remon Oldenbeuving <[email protected]>
Co-authored-by: avtc <[email protected]>
Co-authored-by: Shariq Riaz <[email protected]>
Co-authored-by: sam hoang <[email protected]>
Co-authored-by: Noritaka Kobayashi <[email protected]>
Co-authored-by: R-omk <[email protected]>
Co-authored-by: Chris Hasson <[email protected]>
Co-authored-by: ChuKhaLi <[email protected]>
Co-authored-by: mini2s <[email protected]>

Author: 28 people

.changeset/clean-taxis-feel.md

@@ -0,0 +1,6 @@
+---
+"roo-cline": patch
+---
+
+Add support for `gemini-2.5-flash-preview-05-20` on the Vertex provider
+Add support for `gemini-2.5-flash-preview-05-20:thinking` on the Vertex provider

.changeset/curly-plants-pull.md

@@ -0,0 +1,11 @@
+---
+"roo-cline": patch
+---
+
+New models for the Chutes provider:
+
+- Qwen/Qwen3-235B-A22B
+- Qwen/Qwen3-32B
+- Qwen/Qwen3-30B-A3B
+- Qwen/Qwen3-14B
+- Qwen/Qwen3-8B

.changeset/dark-swans-marry.md

@@ -0,0 +1,5 @@
+---
+"roo-cline": patch
+---
+
+Shows in the UI when the context is intelligently condensed

.changeset/fair-houses-deny.md

@@ -0,0 +1,5 @@
+---
+"roo-cline": patch
+---
+
+Add a UI indicator while the context is condensing

.changeset/four-emus-invite.md

@@ -0,0 +1,5 @@
+---
+"roo-cline": minor
+---
+
+Added an auto-approve API request limit setting similar to Cline

.changeset/fruity-spoons-smash.md

@@ -0,0 +1,5 @@
+---
+"roo-cline": patch
+---
+
+Use YAML as default custom modes format

.changeset/gold-meals-tell.md

@@ -0,0 +1,10 @@
+---
+"roo-cline": minor
+---
+
+Adds refresh models button for Unbound provider
+Adds a button above model picker to refresh models based on the current API Key.
+
+1. Clicking the refresh button saves the API Key and calls /models endpoint using that.
+2. Gets the new models and updates the current model if it is invalid for the given API Key.
+3. The refresh button also flushes existing Unbound models and refetches them.

.changeset/lovely-moles-end.md

@@ -0,0 +1,5 @@
+---
+"roo-cline": patch
+---
+
+Fixed bug that prevented some file links from working in the Agent output