- Add codebaseIndexCustomQueryInstruction field to CodebaseIndexConfig schema
- Update all embedder implementations to accept and use custom query instructions
- Add UI input field in CodeIndexPopover for configuring custom instructions
- Add comprehensive tests for custom query instruction functionality
- Add i18n translations for new UI elements
This allows users to customize the query prefix/instruction for models like
embeddinggemma and qwen3-embedding that benefit from instruction-aware
embeddings.
Fixes#8759
* feat(zai): add GLM-4.5-X, AirX, Flash; sync with Z.ai docs; keep canonical api line keys
* feat(zai): add GLM-4.5V vision model (supportsImages, pricing, 16K max output); add tests
* feat(types,zai): sync Z.AI international model map and tests
- Update pricing, context window, and capabilities for:
glm-4.5-x, glm-4.5-airx, glm-4.5-flash, glm-4.5v, glm-4.6
- Add glm-4-32b-0414-128k
- Align tests with new model specs
* fix(zai): align handler generics with expanded model ids to satisfy CI compile step
* chore(zai): remove tier pricing blocks for Z.ai models
* fix(zai): simplify names in zaiApiLineConfigs for clarity
* chore(zai): set default temperature to 0.6
---------
Co-authored-by: Roo Code <roomote@roocode.com>
chore(gpt5): stop persisting instructions/reasoning_summary in UI message metadata
Problem: ui_messages.json was getting bloated with unused or duplicated content (system 'instructions' and 'reasoning_summary') that we do not read back. Root cause: earlier OpenAI Responses API implementation persisted these fields to per-message metadata; however, 'instructions' are already sent as top-level request instructions and 'reasoning_summary' is surfaced live via streaming events. Neither field is consumed from storage. Changes: (1) Task.persistGpt5Metadata now stores only previous_response_id; (2) removed instructions and reasoning_summary from types; (3) updated Zod schema; (4) persistence layer writes messages as-is (no sanitizer); (5) tests green. Impact: smaller ui_messages.json, no runtime behavior change for requests. Migration: old metadata fields will be ignored by schema.
Fixes#8371 - Updates the model ID from anthropic.claude-4.5-sonnet-v1:0
to anthropic.claude-sonnet-4-5-20250929-v1:0 to match AWS Bedrock naming convention
Co-authored-by: Roo Code <roomote@roocode.com>
* feat: add telemetry tracking to DismissibleUpsell component
- Added UPSELL_DISMISSED and UPSELL_CLICKED events to TelemetryEventName enum
- Updated DismissibleUpsell component to track clicks and dismissals with telemetry
- Added telemetry tests to DismissibleUpsell test suite
- Events include upsellId in the payload for tracking specific upsells
* refactor(webview): make handleDismiss synchronous in DismissibleUpsell
test(webview): add scenario where dismissOnClick=true without onClick tracks only UPSELL_DISMISSED; update tests; all tests passing locally
---------
Co-authored-by: Roo Code <roomote@roocode.com>
* fix: transform keybindings command references in nightly build
The keybindings section was not being transformed during the nightly build process, causing command+y keybinding to reference the wrong command name (roo-cline.addToContext instead of roo-code-nightly.addToContext).
- Added keybindings schema to types.ts
- Updated generatePackageJson to transform keybindings command references
- This ensures keybindings work correctly in the nightly build
* fix: only include keybindings in output when they exist
Updated generatePackageJson to conditionally add keybindings to avoid including undefined values in the generated package.json. Fixed eslint-disable comment placement.
---------
Co-authored-by: Roo Code <roomote@roocode.com>
feat: add organizationFeaturesSchema with roomoteControlEnabled
- Add organizationFeaturesSchema with optional roomoteControlEnabled boolean
- Integrate features property into organizationSettingsSchema as optional
- Add comprehensive tests for new schema validation
- Maintain backward compatibility with existing organization settings
Co-authored-by: Roo Code <roomote@roocode.com>
* feat(cloud): Add telemetry retry queue for network resilience
- Implement RetryQueue class with workspace-scoped persistence
- Queue failed telemetry events for automatic retry
- Retry events every 60 seconds with fresh auth tokens
- FIFO eviction when queue reaches 100 events
- Persist queue across VS Code restarts
This ensures telemetry data isn't lost during network failures or temporary server issues.
Migrated from RooCodeInc/Roo-Code-Cloud#744
* fix: address PR review feedback for retry queue
- Fix retry order to use consistent FIFO processing
- Add retry limit enforcement with max retries check
- Add configurable request timeout (default 30s)
- Add comprehensive tests for retryAll() method
- Add request-max-retries-exceeded event
- Fix timeout test to avoid timing issues
* fix: resolve TypeScript errors in RetryQueue tests
* fix(cloud): Address PR feedback for telemetry retry queue
- Handle HTTP error status codes (500s, 401/403, 429) as failures that trigger retry
- Remove queuing of backfill operations since they're user-initiated
- Fix race condition in concurrent retry processing with isProcessing flag
- Add specialized retry logic for 429 with Retry-After header support
- Clean up unnecessary comments
- Add comprehensive tests for new status code handling
- Add temporary debug logs with emojis for testing
* refactor: address PR feedback for telemetry retry queue
- Remove unused X-Organization-Id header from auth header provider
- Simplify enqueue() API by removing operation parameter
- Fix error retry logic: only retry 5xx, 429, and network failures
- Stop retrying 4xx client errors (400, 401, 403, 404, 422)
- Implement queue-wide pause for 429 rate limiting
- Add auth state management integration:
- Pause queue when not in active-session
- Clear queue on logout or user change
- Preserve queue when same user logs back in
- Remove debug comments
- Fix ESLint no-case-declarations error with proper block scope
- Update tests for all new behaviors
