* feat: add per-project default model override for new features
- Add defaultFeatureModel to ProjectSettings type for project-level override
- Add defaultFeatureModel to Project interface for UI state
- Display Default Feature Model in Model Defaults section alongside phase models
- Include Default Feature Model in global Bulk Replace dialog
- Add Default Feature Model override section to Project Settings
- Add setProjectDefaultFeatureModel store action for project-level overrides
- Update clearAllProjectPhaseModelOverrides to also clear defaultFeatureModel
- Update add-feature-dialog to use project override when available
- Include Default Feature Model in Project Bulk Replace dialog
This allows projects with different complexity levels to use different
default models (e.g., Haiku for simple tasks, Opus for complex projects).
* fix: add server-side __CLEAR__ handler for defaultFeatureModel
- Add handler in settings-service.ts to properly delete defaultFeatureModel
when '__CLEAR__' marker is sent from the UI
- Fix bulk-replace-dialog.tsx to correctly return claude-opus when resetting
default feature model to Anthropic Direct (was incorrectly using
enhancementModel's settings which default to sonnet)
These fixes ensure:
1. Clearing project default model override properly removes the setting
instead of storing literal '__CLEAR__' string
2. Global bulk replace correctly resets default feature model to opus
* fix: include defaultFeatureModel in Reset to Defaults action
- Updated resetPhaseModels to also reset defaultFeatureModel to claude-opus
- Fixed initial state to use canonical 'claude-opus' instead of 'opus'
* refactor: use DEFAULT_GLOBAL_SETTINGS constant for defaultFeatureModel
Address PR review feedback:
- Replace hardcoded { model: 'claude-opus' } with DEFAULT_GLOBAL_SETTINGS.defaultFeatureModel
- Fix Prettier formatting for long destructuring lines
- Import DEFAULT_GLOBAL_SETTINGS from @automaker/types where needed
This improves maintainability by centralizing the default value.
* fix: hide Cursor models in selector when provider is disabled
The Cursor Models section was appearing in model dropdown selectors even
when the Cursor provider was toggled OFF in Settings → AI Providers.
This fix adds a !isCursorDisabled check to the rendering condition,
matching the pattern already used by Codex and OpenCode providers.
This ensures consistency across all provider types.
Fixes the issue where:
- Codex/OpenCode correctly hide models when disabled
- Cursor incorrectly showed models even when disabled
* style: fix Prettier formatting
Fix two related issues with bulk model updates in Kanban view:
1. Bulk update now properly invalidates React Query cache
- Changed handleBulkUpdate and bulk verify handler to call loadFeatures()
- This ensures UI immediately reflects bulk changes
2. Custom provider models (GLM, MiniMax, etc.) now display correctly
- Added fallback lookup in PhaseModelSelector by model ID
- Updated mass-edit-dialog to track providerId after selection
* feat: refactor Claude API Profiles to Claude Compatible Providers
- Rename ClaudeApiProfile to ClaudeCompatibleProvider with models[] array
- Each ProviderModel has mapsToClaudeModel field for Claude tier mapping
- Add providerType field for provider-specific icons (glm, minimax, openrouter)
- Add thinking level support for provider models in phase selectors
- Show all mapped Claude models per provider model (e.g., "Maps to Haiku, Sonnet, Opus")
- Add Bulk Replace feature to switch all phases to a provider at once
- Hide Bulk Replace button when no providers are enabled
- Fix project-level phaseModelOverrides not persisting after refresh
- Fix deleting last provider not persisting (remove empty array guard)
- Add getProviderByModelId() helper for all SDK routes
- Update all routes to pass provider config for provider models
- Update terminology from "profiles" to "providers" throughout UI
- Update documentation to reflect new provider system
* fix: atomic writer race condition and bulk replace reset to defaults
1. AtomicWriter Race Condition Fix (libs/utils/src/atomic-writer.ts):
- Changed temp file naming from Date.now() to Date.now() + random hex
- Uses crypto.randomBytes(4).toString('hex') for uniqueness
- Prevents ENOENT errors when multiple concurrent writes happen
within the same millisecond
2. Bulk Replace "Anthropic Direct" Reset (both dialogs):
- When selecting "Anthropic Direct", now uses DEFAULT_PHASE_MODELS
- Properly resets thinking levels and other settings to defaults
- Added thinkingLevel to the change detection comparison
- Affects both global and project-level bulk replace dialogs
* fix: update tests for new model resolver passthrough behavior
1. model-resolver tests:
- Unknown models now pass through unchanged (provider model support)
- Removed expectations for warnings on unknown models
- Updated case sensitivity and edge case tests accordingly
- Added tests for provider-like model names (GLM-4.7, MiniMax-M2.1)
2. atomic-writer tests:
- Updated regex to match new temp file format with random suffix
- Format changed from .tmp.{timestamp} to .tmp.{timestamp}.{hex}
* refactor: simplify getPhaseModelWithOverrides calls per code review
Address code review feedback on PR #629:
- Make settingsService parameter optional in getPhaseModelWithOverrides
- Function now handles undefined settingsService gracefully by returning defaults
- Remove redundant ternary checks in 4 call sites:
- apps/server/src/routes/context/routes/describe-file.ts
- apps/server/src/routes/context/routes/describe-image.ts
- apps/server/src/routes/worktree/routes/generate-commit-message.ts
- apps/server/src/services/auto-mode-service.ts
- Remove unused DEFAULT_PHASE_MODELS imports where applicable
* test: fix server tests for provider model passthrough behavior
- Update model-resolver.test.ts to expect unknown models to pass through
unchanged (supports ClaudeCompatibleProvider models like GLM-4.7)
- Remove warning expectations for unknown models (valid for providers)
- Add missing getCredentials and getGlobalSettings mocks to
ideation-service.test.ts for settingsService
* fix: address code review feedback for model providers
- Honor thinkingLevel in generate-commit-message.ts
- Pass claudeCompatibleProvider in ideation-service.ts for provider models
- Resolve provider configuration for model overrides in generate-suggestions.ts
- Update "Active Profile" to "Active Provider" label in project-claude-section
- Use substring instead of deprecated substr in api-profiles-section
- Preserve provider enabled state when editing in api-profiles-section
* fix: address CodeRabbit review issues for Claude Compatible Providers
- Fix TypeScript TS2339 error in generate-suggestions.ts where
settingsService was narrowed to 'never' type in else branch
- Use DEFAULT_PHASE_MODELS per-phase defaults instead of hardcoded
'sonnet' in settings-helpers.ts
- Remove duplicate eventHooks key in use-settings-migration.ts
- Add claudeCompatibleProviders to localStorage migration parsing
and merging functions
- Handle canonical claude-* model IDs (claude-haiku, claude-sonnet,
claude-opus) in project-models-section display names
This resolves the CI build failures and addresses code review feedback.
* fix: skip broken list-view-priority E2E test and add Priority column label
- Skip list-view-priority.spec.ts with TODO explaining the infrastructure
issue: setupRealProject only sets localStorage but server settings
take precedence with localStorageMigrated: true
- Add 'Priority' label to list-header.tsx for the priority column
(was empty string, now shows proper header text)
- Increase column width to accommodate the label
The E2E test issue is that tests create features in a temp directory,
but the server loads from the E2E Test Project fixture path set in
setup-e2e-fixtures.mjs. Needs infrastructure fix to properly switch
projects or create features through UI instead of on disk.
This commit introduces significant updates to the cursor model handling and auto mode features. The cursor model IDs have been standardized to a canonical format, ensuring backward compatibility while migrating legacy IDs. New endpoints for starting and stopping the auto mode loop have been added, allowing for better control over project-specific auto mode operations.
Key changes:
- Updated cursor model IDs to use the 'cursor-' prefix for consistency.
- Added new API endpoints: `/start` and `/stop` for managing auto mode.
- Enhanced the status endpoint to provide detailed project-specific auto mode information.
- Improved error handling and logging throughout the auto mode service.
- Migrated legacy model IDs to their canonical counterparts in various components.
This update aims to streamline the user experience and ensure a smooth transition for existing users while providing new functionalities.
The phase model selector was showing ALL discovered dynamic models regardless
of whether they were enabled in settings. Now it filters dynamic models by
enabledDynamicModelIds, matching the behavior of Cursor models and making
the enable/disable setting meaningful.
Users can now:
- Disable models in settings they don't want to use
- See only enabled dynamic models in the model selector dropdown
- Have the "Select all" checkbox properly control which models appear
This ensures consistency: enabling/disabling models in settings affects
which models are available for feature execution.
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
- Added functionality to save, clear, and load backlog plans within the application.
- Introduced a new API endpoint for clearing saved backlog plans.
- Enhanced the backlog plan dialog to allow users to review and apply changes to their features.
- Integrated dependency management features in the UI, allowing users to select parent and child dependencies for features.
- Improved the graph view with options to manage plans and visualize dependencies effectively.
- Updated the sidebar and settings to include provider visibility toggles for better user control over model selection.
These changes aim to enhance the user experience by providing robust backlog management capabilities and improving the overall UI for feature planning.
Problem:
- OpenRouter dynamic models only appeared after visiting settings page
- PhaseModelSelector (used in Add/Edit Feature dialogs) only fetched Codex models
- dynamicOpencodeModels remained empty until OpencodeSettingsTab mounted
Solution:
- Add fetchOpencodeModels() action to app-store mirroring fetchCodexModels pattern
- Add state tracking: opencodeModelsLoading, opencodeModelsError, timestamps
- Call fetchOpencodeModels() in PhaseModelSelector useEffect on mount
- Use same caching strategy: 5min success cache, 30sec failure cooldown
Files changed:
- apps/ui/src/store/app-store.ts
- Add OpenCode model loading state properties
- Add fetchOpencodeModels action with error handling & caching
- apps/ui/src/components/views/settings-view/model-defaults/phase-model-selector.tsx
- Add opencodeModelsLoading, fetchOpencodeModels to store hook
- Add useEffect to fetch OpenCode models on mount
Result:
- OpenRouter models now appear in Add/Edit Feature dialogs immediately
- No need to visit settings page first
- Consistent with Codex model loading behavior
- Implemented a new endpoint to generate commit messages based on git diffs.
- Updated worktree routes to include the AI commit message generation functionality.
- Enhanced the UI to support automatic generation of commit messages when the commit dialog opens, based on user settings.
- Added settings for enabling/disabling AI-generated commit messages and configuring the model used for generation.
- Update isOpencodeModel() to detect dynamic models with provider/model format
(e.g., github-copilot/gpt-4o, google/gemini-2.5-pro, zai-coding-plan/glm-4.7)
- Update resolveModelString() to recognize and pass through OpenCode models
- Update enhance route to route OpenCode models to OpenCode provider
- Fix OpenCode CLI command format: use --format json (not stream-json)
- Remove unsupported -q and - flags from CLI arguments
- Update normalizeEvent() to handle actual OpenCode JSON event format
- Add dynamic model configuration UI with provider grouping
- Cache providers and models in app store for snappier navigation
- Show authenticated providers in OpenCode CLI status card
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add official MiniMax logo SVG from LobeHub icons library
- Add official Z.ai logo SVG for GLM models from LobeHub icons library
- Add BigPickle icon with custom green color (#4ADE80)
- Fix icon detection logic to properly handle amazon-bedrock/ and opencode/ prefixes
- Update phase-model-selector and opencode-model-configuration to use
getProviderIconForModel() for consistent icon display across the app
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The issue was that ALL OpenCode models were showing the OpenCode icon, regardless
of their actual underlying provider. This fix ensures each model shows its
authentic brand icon.
Changes:
1. **model-constants.ts** - Fixed provider field assignment
- Changed provider from hardcoded 'opencode' to actual config.provider
- Now correctly maps: opencode/big-pickle, amazon-bedrock/anthropic.*, etc.
2. **phase-model-selector.tsx** - Added provider-specific icon logic
- Added imports for DeepSeekIcon, NovaIcon, QwenIcon, MistralIcon, MetaIcon
- Added ProviderIcon selector based on model.provider field
- Each model type now displays its correct provider icon
3. **provider-icon.tsx** - Updated icon detection and mapping
- Enhanced getUnderlyingModelIcon() to detect specific Bedrock providers:
* amazon-bedrock/anthropic.* → anthropic icon
* amazon-bedrock/deepseek.* → deepseek icon
* amazon-bedrock/nova.* → nova icon
* amazon-bedrock/meta.* or llama → meta icon
* amazon-bedrock/mistral.* → mistral icon
* amazon-bedrock/qwen.* → qwen icon
* opencode/* → opencode icon
- Added meta and mistral to PROVIDER_ICON_KEYS
- Added placeholder definitions for meta/mistral in PROVIDER_ICON_DEFINITIONS
- Updated iconMap to include all provider icons
- Set OpenCode icon to official brand color (#6366F1 indigo)
Result: All model selectors and kanban cards now show correct brand icons
for each OpenCode model (DeepSeek whale, Amazon Nova sparkle, Qwen star, etc.)
- Bumped version numbers for @automaker/server and @automaker/ui to 0.9.0 in package-lock.json.
- Introduced CodexAppServerService and CodexModelCacheService to manage communication with the Codex CLI's app-server and cache model data.
- Updated CodexUsageService to utilize app-server for fetching usage data.
- Enhanced Codex routes to support fetching available models and integrated model caching.
- Improved UI components to dynamically load and display Codex models, including error handling and loading states.
- Added new API methods for fetching Codex models and integrated them into the app store for state management.
These changes improve the overall functionality and user experience of the Codex integration, ensuring efficient model management and data retrieval.
- Implemented a collapsible dropdown for navigation items in the settings view, allowing users to expand or collapse sub-items.
- Added local storage functionality to remember the open/closed state of the dropdown across sessions.
- Updated the styling and interaction logic for improved user experience and accessibility.
These changes improve the organization of navigation items and enhance user interaction within the settings view.
* memory
* feat: add smart memory selection with task context
- Add taskContext parameter to loadContextFiles for intelligent file selection
- Memory files are scored based on tag matching with task keywords
- Category name matching (e.g., "terminals" matches terminals.md) with 4x weight
- Usage statistics influence scoring (files that helped before rank higher)
- Limit to top 5 files + always include gotchas.md
- Auto-mode passes feature title/description as context
- Chat sessions pass user message as context
This prevents loading 40+ memory files and killing context limits.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
* refactor: enhance auto-mode service and context loader
- Improved context loading by adding task context for better memory selection.
- Updated JSON parsing logic to handle various formats and ensure robust error handling.
- Introduced file locking mechanisms to prevent race conditions during memory file updates.
- Enhanced metadata handling in memory files, including validation and sanitization.
- Refactored scoring logic for context files to improve selection accuracy based on task relevance.
These changes optimize memory file management and enhance the overall performance of the auto-mode service.
* refactor: enhance learning extraction and formatting in auto-mode service
- Improved the learning extraction process by refining the user prompt to focus on meaningful insights and structured JSON output.
- Updated the LearningEntry interface to include additional context fields for better documentation of decisions and patterns.
- Enhanced the formatLearning function to adopt an Architecture Decision Record (ADR) style, providing richer context for recorded learnings.
- Added detailed logging for better traceability during the learning extraction and appending processes.
These changes aim to improve the quality and clarity of learnings captured during the auto-mode service's operation.
* feat: integrate stripProviderPrefix utility for model ID handling
- Added stripProviderPrefix utility to various routes to ensure providers receive bare model IDs.
- Updated model references in executeQuery calls across multiple files, enhancing consistency in model ID handling.
- Introduced memoryExtractionModel in settings for improved learning extraction tasks.
These changes streamline the model ID processing and enhance the overall functionality of the provider interactions.
* feat: enhance error handling and server offline management in board actions
- Improved error handling in the handleRunFeature and handleStartImplementation functions to throw errors for better caller management.
- Integrated connection error detection and server offline handling, redirecting users to the login page when the server is unreachable.
- Updated follow-up feature logic to include rollback mechanisms and improved user feedback for error scenarios.
These changes enhance the robustness of the board actions by ensuring proper error management and user experience during server connectivity issues.
---------
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: webdevcody <webdevcody@gmail.com>
- Added OpenCode model selection to AddFeatureDialog and EditFeatureDialog.
- Introduced ProfileTypeahead component for improved profile selection.
- Updated model constants to include OpenCode models and integrated them into the PhaseModelSelector.
- Enhanced planning mode options with new UI elements for OpenCode.
- Refactored existing components to streamline model handling and improve user experience.
This commit expands the functionality of the feature dialogs, allowing users to select and manage OpenCode models effectively.
- Add ReasoningEffortSelector component for UI selection
- Integrate reasoning effort in feature creation/editing dialogs
- Add reasoning effort support to phase model selector
- Update agent service and board actions to handle reasoning effort
- Add reasoning effort fields to feature and settings types
- Update model selector and agent info panel with reasoning effort display
- Enhance agent context parser for reasoning effort processing
Reasoning effort allows fine-tuned control over Codex model reasoning
capabilities, providing options from 'none' to 'xhigh' for different
task complexity requirements.
- Migrated settings persistence from localStorage to an API-first approach, ensuring consistency between Electron and web modes.
- Introduced `useSettingsSync` hook for automatic synchronization of settings to the server with debouncing.
- Enhanced feature update logic to track description changes with a history, allowing for better management of feature descriptions.
- Updated various components and services to utilize the new settings structure and description history functionality.
- Removed persist middleware from Zustand store, streamlining state management and improving performance.
- Updated imports and references from 'phase-models' to 'model-defaults' across various components.
- Removed obsolete phase models index file to streamline the codebase.
