lint

Co-authored-by: -LAN- <laipz8200@outlook.com>

.claude/settings.json

@@ -0,0 +1,8 @@
+{
+  "enabledPlugins": {
+    "feature-dev@claude-plugins-official": true,
+    "context7@claude-plugins-official": true,
+    "typescript-lsp@claude-plugins-official": true,
+    "pyright-lsp@claude-plugins-official": true
+  }
+}

.claude/settings.json.example

@@ -1,19 +0,0 @@
-{
-    "permissions": {
-      "allow": [],
-      "deny": []
-    },
-    "env": {
-      "__comment": "Environment variables for MCP servers. Override in .claude/settings.local.json with actual values.",
-      "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
-    },
-    "enabledMcpjsonServers": [
-      "context7",
-      "sequential-thinking",
-      "github",
-      "fetch",
-      "playwright",
-      "ide"
-    ],
-    "enableAllProjectMcpServers": true
-  }

.claude/skills/component-refactoring/SKILL.md

@@ -0,0 +1,483 @@
+---
+name: component-refactoring
+description: Refactor high-complexity React components in Dify frontend. Use when `pnpm analyze-component --json` shows complexity > 50 or lineCount > 300, when the user asks for code splitting, hook extraction, or complexity reduction, or when `pnpm analyze-component` warns to refactor before testing; avoid for simple/well-structured components, third-party wrappers, or when the user explicitly wants testing without refactoring.
+---
+
+# Dify Component Refactoring Skill
+
+Refactor high-complexity React components in the Dify frontend codebase with the patterns and workflow below.
+
+> **Complexity Threshold**: Components with complexity > 50 (measured by `pnpm analyze-component`) should be refactored before testing.
+
+## Quick Reference
+
+### Commands (run from `web/`)
+
+Use paths relative to `web/` (e.g., `app/components/...`).
+Use `refactor-component` for refactoring prompts and `analyze-component` for testing prompts and metrics.
+
+```bash
+cd web
+
+# Generate refactoring prompt
+pnpm refactor-component <path>
+
+# Output refactoring analysis as JSON
+pnpm refactor-component <path> --json
+
+# Generate testing prompt (after refactoring)
+pnpm analyze-component <path>
+
+# Output testing analysis as JSON
+pnpm analyze-component <path> --json
+```
+
+### Complexity Analysis
+
+```bash
+# Analyze component complexity
+pnpm analyze-component <path> --json
+
+# Key metrics to check:
+# - complexity: normalized score 0-100 (target < 50)
+# - maxComplexity: highest single function complexity
+# - lineCount: total lines (target < 300)
+```
+
+### Complexity Score Interpretation
+
+| Score | Level | Action |
+|-------|-------|--------|
+| 0-25 | 🟢 Simple | Ready for testing |
+| 26-50 | 🟡 Medium | Consider minor refactoring |
+| 51-75 | 🟠 Complex | **Refactor before testing** |
+| 76-100 | 🔴 Very Complex | **Must refactor** |
+
+## Core Refactoring Patterns
+
+### Pattern 1: Extract Custom Hooks
+
+**When**: Component has complex state management, multiple `useState`/`useEffect`, or business logic mixed with UI.
+
+**Dify Convention**: Place hooks in a `hooks/` subdirectory or alongside the component as `use-<feature>.ts`.
+
+```typescript
+// ❌ Before: Complex state logic in component
+const Configuration: FC = () => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+  const [datasetConfigs, setDatasetConfigs] = useState<DatasetConfigs>(...)
+  const [completionParams, setCompletionParams] = useState<FormValue>({})
+  
+  // 50+ lines of state management logic...
+  
+  return <div>...</div>
+}
+
+// ✅ After: Extract to custom hook
+// hooks/use-model-config.ts
+export const useModelConfig = (appId: string) => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+  const [completionParams, setCompletionParams] = useState<FormValue>({})
+  
+  // Related state management logic here
+  
+  return { modelConfig, setModelConfig, completionParams, setCompletionParams }
+}
+
+// Component becomes cleaner
+const Configuration: FC = () => {
+  const { modelConfig, setModelConfig } = useModelConfig(appId)
+  return <div>...</div>
+}
+```
+
+**Dify Examples**:
+- `web/app/components/app/configuration/hooks/use-advanced-prompt-config.ts`
+- `web/app/components/app/configuration/debug/hooks.tsx`
+- `web/app/components/workflow/hooks/use-workflow.ts`
+
+### Pattern 2: Extract Sub-Components
+
+**When**: Single component has multiple UI sections, conditional rendering blocks, or repeated patterns.
+
+**Dify Convention**: Place sub-components in subdirectories or as separate files in the same directory.
+
+```typescript
+// ❌ Before: Monolithic JSX with multiple sections
+const AppInfo = () => {
+  return (
+    <div>
+      {/* 100 lines of header UI */}
+      {/* 100 lines of operations UI */}
+      {/* 100 lines of modals */}
+    </div>
+  )
+}
+
+// ✅ After: Split into focused components
+// app-info/
+//   ├── index.tsx           (orchestration only)
+//   ├── app-header.tsx      (header UI)
+//   ├── app-operations.tsx  (operations UI)
+//   └── app-modals.tsx      (modal management)
+
+const AppInfo = () => {
+  const { showModal, setShowModal } = useAppInfoModals()
+  
+  return (
+    <div>
+      <AppHeader appDetail={appDetail} />
+      <AppOperations onAction={handleAction} />
+      <AppModals show={showModal} onClose={() => setShowModal(null)} />
+    </div>
+  )
+}
+```
+
+**Dify Examples**:
+- `web/app/components/app/configuration/` directory structure
+- `web/app/components/workflow/nodes/` per-node organization
+
+### Pattern 3: Simplify Conditional Logic
+
+**When**: Deep nesting (> 3 levels), complex ternaries, or multiple `if/else` chains.
+
+```typescript
+// ❌ Before: Deeply nested conditionals
+const Template = useMemo(() => {
+  if (appDetail?.mode === AppModeEnum.CHAT) {
+    switch (locale) {
+      case LanguagesSupported[1]:
+        return <TemplateChatZh />
+      case LanguagesSupported[7]:
+        return <TemplateChatJa />
+      default:
+        return <TemplateChatEn />
+    }
+  }
+  if (appDetail?.mode === AppModeEnum.ADVANCED_CHAT) {
+    // Another 15 lines...
+  }
+  // More conditions...
+}, [appDetail, locale])
+
+// ✅ After: Use lookup tables + early returns
+const TEMPLATE_MAP = {
+  [AppModeEnum.CHAT]: {
+    [LanguagesSupported[1]]: TemplateChatZh,
+    [LanguagesSupported[7]]: TemplateChatJa,
+    default: TemplateChatEn,
+  },
+  [AppModeEnum.ADVANCED_CHAT]: {
+    [LanguagesSupported[1]]: TemplateAdvancedChatZh,
+    // ...
+  },
+}
+
+const Template = useMemo(() => {
+  const modeTemplates = TEMPLATE_MAP[appDetail?.mode]
+  if (!modeTemplates) return null
+  
+  const TemplateComponent = modeTemplates[locale] || modeTemplates.default
+  return <TemplateComponent appDetail={appDetail} />
+}, [appDetail, locale])
+```
+
+### Pattern 4: Extract API/Data Logic
+
+**When**: Component directly handles API calls, data transformation, or complex async operations.
+
+**Dify Convention**: Use `@tanstack/react-query` hooks from `web/service/use-*.ts` or create custom data hooks.
+
+```typescript
+// ❌ Before: API logic in component
+const MCPServiceCard = () => {
+  const [basicAppConfig, setBasicAppConfig] = useState({})
+  
+  useEffect(() => {
+    if (isBasicApp && appId) {
+      (async () => {
+        const res = await fetchAppDetail({ url: '/apps', id: appId })
+        setBasicAppConfig(res?.model_config || {})
+      })()
+    }
+  }, [appId, isBasicApp])
+  
+  // More API-related logic...
+}
+
+// ✅ After: Extract to data hook using React Query
+// use-app-config.ts
+import { useQuery } from '@tanstack/react-query'
+import { get } from '@/service/base'
+
+const NAME_SPACE = 'appConfig'
+
+export const useAppConfig = (appId: string, isBasicApp: boolean) => {
+  return useQuery({
+    enabled: isBasicApp && !!appId,
+    queryKey: [NAME_SPACE, 'detail', appId],
+    queryFn: () => get<AppDetailResponse>(`/apps/${appId}`),
+    select: data => data?.model_config || {},
+  })
+}
+
+// Component becomes cleaner
+const MCPServiceCard = () => {
+  const { data: config, isLoading } = useAppConfig(appId, isBasicApp)
+  // UI only
+}
+```
+
+**React Query Best Practices in Dify**:
+- Define `NAME_SPACE` for query key organization
+- Use `enabled` option for conditional fetching
+- Use `select` for data transformation
+- Export invalidation hooks: `useInvalidXxx`
+
+**Dify Examples**:
+- `web/service/use-workflow.ts`
+- `web/service/use-common.ts`
+- `web/service/knowledge/use-dataset.ts`
+- `web/service/knowledge/use-document.ts`
+
+### Pattern 5: Extract Modal/Dialog Management
+
+**When**: Component manages multiple modals with complex open/close states.
+
+**Dify Convention**: Modals should be extracted with their state management.
+
+```typescript
+// ❌ Before: Multiple modal states in component
+const AppInfo = () => {
+  const [showEditModal, setShowEditModal] = useState(false)
+  const [showDuplicateModal, setShowDuplicateModal] = useState(false)
+  const [showConfirmDelete, setShowConfirmDelete] = useState(false)
+  const [showSwitchModal, setShowSwitchModal] = useState(false)
+  const [showImportDSLModal, setShowImportDSLModal] = useState(false)
+  // 5+ more modal states...
+}
+
+// ✅ After: Extract to modal management hook
+type ModalType = 'edit' | 'duplicate' | 'delete' | 'switch' | 'import' | null
+
+const useAppInfoModals = () => {
+  const [activeModal, setActiveModal] = useState<ModalType>(null)
+  
+  const openModal = useCallback((type: ModalType) => setActiveModal(type), [])
+  const closeModal = useCallback(() => setActiveModal(null), [])
+  
+  return {
+    activeModal,
+    openModal,
+    closeModal,
+    isOpen: (type: ModalType) => activeModal === type,
+  }
+}
+```
+
+### Pattern 6: Extract Form Logic
+
+**When**: Complex form validation, submission handling, or field transformation.
+
+**Dify Convention**: Use `@tanstack/react-form` patterns from `web/app/components/base/form/`.
+
+```typescript
+// ✅ Use existing form infrastructure
+import { useAppForm } from '@/app/components/base/form'
+
+const ConfigForm = () => {
+  const form = useAppForm({
+    defaultValues: { name: '', description: '' },
+    onSubmit: handleSubmit,
+  })
+  
+  return <form.Provider>...</form.Provider>
+}
+```
+
+## Dify-Specific Refactoring Guidelines
+
+### 1. Context Provider Extraction
+
+**When**: Component provides complex context values with multiple states.
+
+```typescript
+// ❌ Before: Large context value object
+const value = {
+  appId, isAPIKeySet, isTrailFinished, mode, modelModeType,
+  promptMode, isAdvancedMode, isAgent, isOpenAI, isFunctionCall,
+  // 50+ more properties...
+}
+return <ConfigContext.Provider value={value}>...</ConfigContext.Provider>
+
+// ✅ After: Split into domain-specific contexts
+<ModelConfigProvider value={modelConfigValue}>
+  <DatasetConfigProvider value={datasetConfigValue}>
+    <UIConfigProvider value={uiConfigValue}>
+      {children}
+    </UIConfigProvider>
+  </DatasetConfigProvider>
+</ModelConfigProvider>
+```
+
+**Dify Reference**: `web/context/` directory structure
+
+### 2. Workflow Node Components
+
+**When**: Refactoring workflow node components (`web/app/components/workflow/nodes/`).
+
+**Conventions**:
+- Keep node logic in `use-interactions.ts`
+- Extract panel UI to separate files
+- Use `_base` components for common patterns
+
+```
+nodes/<node-type>/
+  ├── index.tsx              # Node registration
+  ├── node.tsx               # Node visual component
+  ├── panel.tsx              # Configuration panel
+  ├── use-interactions.ts    # Node-specific hooks
+  └── types.ts               # Type definitions
+```
+
+### 3. Configuration Components
+
+**When**: Refactoring app configuration components.
+
+**Conventions**:
+- Separate config sections into subdirectories
+- Use existing patterns from `web/app/components/app/configuration/`
+- Keep feature toggles in dedicated components
+
+### 4. Tool/Plugin Components
+
+**When**: Refactoring tool-related components (`web/app/components/tools/`).
+
+**Conventions**:
+- Follow existing modal patterns
+- Use service hooks from `web/service/use-tools.ts`
+- Keep provider-specific logic isolated
+
+## Refactoring Workflow
+
+### Step 1: Generate Refactoring Prompt
+
+```bash
+pnpm refactor-component <path>
+```
+
+This command will:
+- Analyze component complexity and features
+- Identify specific refactoring actions needed
+- Generate a prompt for AI assistant (auto-copied to clipboard on macOS)
+- Provide detailed requirements based on detected patterns
+
+### Step 2: Analyze Details
+
+```bash
+pnpm analyze-component <path> --json
+```
+
+Identify:
+- Total complexity score
+- Max function complexity
+- Line count
+- Features detected (state, effects, API, etc.)
+
+### Step 3: Plan
+
+Create a refactoring plan based on detected features:
+
+| Detected Feature | Refactoring Action |
+|------------------|-------------------|
+| `hasState: true` + `hasEffects: true` | Extract custom hook |
+| `hasAPI: true` | Extract data/service hook |
+| `hasEvents: true` (many) | Extract event handlers |
+| `lineCount > 300` | Split into sub-components |
+| `maxComplexity > 50` | Simplify conditional logic |
+
+### Step 4: Execute Incrementally
+
+1. **Extract one piece at a time**
+2. **Run lint, type-check, and tests after each extraction**
+3. **Verify functionality before next step**
+
+```
+For each extraction:
+  ┌────────────────────────────────────────┐
+  │ 1. Extract code                        │
+  │ 2. Run: pnpm lint:fix                  │
+  │ 3. Run: pnpm type-check:tsgo           │
+  │ 4. Run: pnpm test                      │
+  │ 5. Test functionality manually         │
+  │ 6. PASS? → Next extraction             │
+  │    FAIL? → Fix before continuing       │
+  └────────────────────────────────────────┘
+```
+
+### Step 5: Verify
+
+After refactoring:
+
+```bash
+# Re-run refactor command to verify improvements
+pnpm refactor-component <path>
+
+# If complexity < 25 and lines < 200, you'll see:
+# ✅ COMPONENT IS WELL-STRUCTURED
+
+# For detailed metrics:
+pnpm analyze-component <path> --json
+
+# Target metrics:
+# - complexity < 50
+# - lineCount < 300
+# - maxComplexity < 30
+```
+
+## Common Mistakes to Avoid
+
+### ❌ Over-Engineering
+
+```typescript
+// ❌ Too many tiny hooks
+const useButtonText = () => useState('Click')
+const useButtonDisabled = () => useState(false)
+const useButtonLoading = () => useState(false)
+
+// ✅ Cohesive hook with related state
+const useButtonState = () => {
+  const [text, setText] = useState('Click')
+  const [disabled, setDisabled] = useState(false)
+  const [loading, setLoading] = useState(false)
+  return { text, setText, disabled, setDisabled, loading, setLoading }
+}
+```
+
+### ❌ Breaking Existing Patterns
+
+- Follow existing directory structures
+- Maintain naming conventions
+- Preserve export patterns for compatibility
+
+### ❌ Premature Abstraction
+
+- Only extract when there's clear complexity benefit
+- Don't create abstractions for single-use code
+- Keep refactored code in the same domain area
+
+## References
+
+### Dify Codebase Examples
+
+- **Hook extraction**: `web/app/components/app/configuration/hooks/`
+- **Component splitting**: `web/app/components/app/configuration/`
+- **Service hooks**: `web/service/use-*.ts`
+- **Workflow patterns**: `web/app/components/workflow/hooks/`
+- **Form patterns**: `web/app/components/base/form/`
+
+### Related Skills
+
+- `frontend-testing` - For testing refactored components
+- `web/testing/testing.md` - Testing specification

.claude/skills/component-refactoring/references/complexity-patterns.md

@@ -0,0 +1,493 @@
+# Complexity Reduction Patterns
+
+This document provides patterns for reducing cognitive complexity in Dify React components.
+
+## Understanding Complexity
+
+### SonarJS Cognitive Complexity
+
+The `pnpm analyze-component` tool uses SonarJS cognitive complexity metrics:
+
+- **Total Complexity**: Sum of all functions' complexity in the file
+- **Max Complexity**: Highest single function complexity
+
+### What Increases Complexity
+
+| Pattern | Complexity Impact |
+|---------|-------------------|
+| `if/else` | +1 per branch |
+| Nested conditions | +1 per nesting level |
+| `switch/case` | +1 per case |
+| `for/while/do` | +1 per loop |
+| `&&`/`||` chains | +1 per operator |
+| Nested callbacks | +1 per nesting level |
+| `try/catch` | +1 per catch |
+| Ternary expressions | +1 per nesting |
+
+## Pattern 1: Replace Conditionals with Lookup Tables
+
+**Before** (complexity: ~15):
+
+```typescript
+const Template = useMemo(() => {
+  if (appDetail?.mode === AppModeEnum.CHAT) {
+    switch (locale) {
+      case LanguagesSupported[1]:
+        return <TemplateChatZh appDetail={appDetail} />
+      case LanguagesSupported[7]:
+        return <TemplateChatJa appDetail={appDetail} />
+      default:
+        return <TemplateChatEn appDetail={appDetail} />
+    }
+  }
+  if (appDetail?.mode === AppModeEnum.ADVANCED_CHAT) {
+    switch (locale) {
+      case LanguagesSupported[1]:
+        return <TemplateAdvancedChatZh appDetail={appDetail} />
+      case LanguagesSupported[7]:
+        return <TemplateAdvancedChatJa appDetail={appDetail} />
+      default:
+        return <TemplateAdvancedChatEn appDetail={appDetail} />
+    }
+  }
+  if (appDetail?.mode === AppModeEnum.WORKFLOW) {
+    // Similar pattern...
+  }
+  return null
+}, [appDetail, locale])
+```
+
+**After** (complexity: ~3):
+
+```typescript
+// Define lookup table outside component
+const TEMPLATE_MAP: Record<AppModeEnum, Record<string, FC<TemplateProps>>> = {
+  [AppModeEnum.CHAT]: {
+    [LanguagesSupported[1]]: TemplateChatZh,
+    [LanguagesSupported[7]]: TemplateChatJa,
+    default: TemplateChatEn,
+  },
+  [AppModeEnum.ADVANCED_CHAT]: {
+    [LanguagesSupported[1]]: TemplateAdvancedChatZh,
+    [LanguagesSupported[7]]: TemplateAdvancedChatJa,
+    default: TemplateAdvancedChatEn,
+  },
+  [AppModeEnum.WORKFLOW]: {
+    [LanguagesSupported[1]]: TemplateWorkflowZh,
+    [LanguagesSupported[7]]: TemplateWorkflowJa,
+    default: TemplateWorkflowEn,
+  },
+  // ...
+}
+
+// Clean component logic
+const Template = useMemo(() => {
+  if (!appDetail?.mode) return null
+  
+  const templates = TEMPLATE_MAP[appDetail.mode]
+  if (!templates) return null
+  
+  const TemplateComponent = templates[locale] ?? templates.default
+  return <TemplateComponent appDetail={appDetail} />
+}, [appDetail, locale])
+```
+
+## Pattern 2: Use Early Returns
+
+**Before** (complexity: ~10):
+
+```typescript
+const handleSubmit = () => {
+  if (isValid) {
+    if (hasChanges) {
+      if (isConnected) {
+        submitData()
+      } else {
+        showConnectionError()
+      }
+    } else {
+      showNoChangesMessage()
+    }
+  } else {
+    showValidationError()
+  }
+}
+```
+
+**After** (complexity: ~4):
+
+```typescript
+const handleSubmit = () => {
+  if (!isValid) {
+    showValidationError()
+    return
+  }
+  
+  if (!hasChanges) {
+    showNoChangesMessage()
+    return
+  }
+  
+  if (!isConnected) {
+    showConnectionError()
+    return
+  }
+  
+  submitData()
+}
+```
+
+## Pattern 3: Extract Complex Conditions
+
+**Before** (complexity: high):
+
+```typescript
+const canPublish = (() => {
+  if (mode !== AppModeEnum.COMPLETION) {
+    if (!isAdvancedMode)
+      return true
+
+    if (modelModeType === ModelModeType.completion) {
+      if (!hasSetBlockStatus.history || !hasSetBlockStatus.query)
+        return false
+      return true
+    }
+    return true
+  }
+  return !promptEmpty
+})()
+```
+
+**After** (complexity: lower):
+
+```typescript
+// Extract to named functions
+const canPublishInCompletionMode = () => !promptEmpty
+
+const canPublishInChatMode = () => {
+  if (!isAdvancedMode) return true
+  if (modelModeType !== ModelModeType.completion) return true
+  return hasSetBlockStatus.history && hasSetBlockStatus.query
+}
+
+// Clean main logic
+const canPublish = mode === AppModeEnum.COMPLETION
+  ? canPublishInCompletionMode()
+  : canPublishInChatMode()
+```
+
+## Pattern 4: Replace Chained Ternaries
+
+**Before** (complexity: ~5):
+
+```typescript
+const statusText = serverActivated
+  ? t('status.running')
+  : serverPublished
+    ? t('status.inactive')
+    : appUnpublished
+      ? t('status.unpublished')
+      : t('status.notConfigured')
+```
+
+**After** (complexity: ~2):
+
+```typescript
+const getStatusText = () => {
+  if (serverActivated) return t('status.running')
+  if (serverPublished) return t('status.inactive')
+  if (appUnpublished) return t('status.unpublished')
+  return t('status.notConfigured')
+}
+
+const statusText = getStatusText()
+```
+
+Or use lookup:
+
+```typescript
+const STATUS_TEXT_MAP = {
+  running: 'status.running',
+  inactive: 'status.inactive',
+  unpublished: 'status.unpublished',
+  notConfigured: 'status.notConfigured',
+} as const
+
+const getStatusKey = (): keyof typeof STATUS_TEXT_MAP => {
+  if (serverActivated) return 'running'
+  if (serverPublished) return 'inactive'
+  if (appUnpublished) return 'unpublished'
+  return 'notConfigured'
+}
+
+const statusText = t(STATUS_TEXT_MAP[getStatusKey()])
+```
+
+## Pattern 5: Flatten Nested Loops
+
+**Before** (complexity: high):
+
+```typescript
+const processData = (items: Item[]) => {
+  const results: ProcessedItem[] = []
+  
+  for (const item of items) {
+    if (item.isValid) {
+      for (const child of item.children) {
+        if (child.isActive) {
+          for (const prop of child.properties) {
+            if (prop.value !== null) {
+              results.push({
+                itemId: item.id,
+                childId: child.id,
+                propValue: prop.value,
+              })
+            }
+          }
+        }
+      }
+    }
+  }
+  
+  return results
+}
+```
+
+**After** (complexity: lower):
+
+```typescript
+// Use functional approach
+const processData = (items: Item[]) => {
+  return items
+    .filter(item => item.isValid)
+    .flatMap(item =>
+      item.children
+        .filter(child => child.isActive)
+        .flatMap(child =>
+          child.properties
+            .filter(prop => prop.value !== null)
+            .map(prop => ({
+              itemId: item.id,
+              childId: child.id,
+              propValue: prop.value,
+            }))
+        )
+    )
+}
+```
+
+## Pattern 6: Extract Event Handler Logic
+
+**Before** (complexity: high in component):
+
+```typescript
+const Component = () => {
+  const handleSelect = (data: DataSet[]) => {
+    if (isEqual(data.map(item => item.id), dataSets.map(item => item.id))) {
+      hideSelectDataSet()
+      return
+    }
+
+    formattingChangedDispatcher()
+    let newDatasets = data
+    if (data.find(item => !item.name)) {
+      const newSelected = produce(data, (draft) => {
+        data.forEach((item, index) => {
+          if (!item.name) {
+            const newItem = dataSets.find(i => i.id === item.id)
+            if (newItem)
+              draft[index] = newItem
+          }
+        })
+      })
+      setDataSets(newSelected)
+      newDatasets = newSelected
+    }
+    else {
+      setDataSets(data)
+    }
+    hideSelectDataSet()
+    
+    // 40 more lines of logic...
+  }
+  
+  return <div>...</div>
+}
+```
+
+**After** (complexity: lower):
+
+```typescript
+// Extract to hook or utility
+const useDatasetSelection = (dataSets: DataSet[], setDataSets: SetState<DataSet[]>) => {
+  const normalizeSelection = (data: DataSet[]) => {
+    const hasUnloadedItem = data.some(item => !item.name)
+    if (!hasUnloadedItem) return data
+    
+    return produce(data, (draft) => {
+      data.forEach((item, index) => {
+        if (!item.name) {
+          const existing = dataSets.find(i => i.id === item.id)
+          if (existing) draft[index] = existing
+        }
+      })
+    })
+  }
+  
+  const hasSelectionChanged = (newData: DataSet[]) => {
+    return !isEqual(
+      newData.map(item => item.id),
+      dataSets.map(item => item.id)
+    )
+  }
+  
+  return { normalizeSelection, hasSelectionChanged }
+}
+
+// Component becomes cleaner
+const Component = () => {
+  const { normalizeSelection, hasSelectionChanged } = useDatasetSelection(dataSets, setDataSets)
+  
+  const handleSelect = (data: DataSet[]) => {
+    if (!hasSelectionChanged(data)) {
+      hideSelectDataSet()
+      return
+    }
+    
+    formattingChangedDispatcher()
+    const normalized = normalizeSelection(data)
+    setDataSets(normalized)
+    hideSelectDataSet()
+  }
+  
+  return <div>...</div>
+}
+```
+
+## Pattern 7: Reduce Boolean Logic Complexity
+
+**Before** (complexity: ~8):
+
+```typescript
+const toggleDisabled = hasInsufficientPermissions
+  || appUnpublished
+  || missingStartNode
+  || triggerModeDisabled
+  || (isAdvancedApp && !currentWorkflow?.graph)
+  || (isBasicApp && !basicAppConfig.updated_at)
+```
+
+**After** (complexity: ~3):
+
+```typescript
+// Extract meaningful boolean functions
+const isAppReady = () => {
+  if (isAdvancedApp) return !!currentWorkflow?.graph
+  return !!basicAppConfig.updated_at
+}
+
+const hasRequiredPermissions = () => {
+  return isCurrentWorkspaceEditor && !hasInsufficientPermissions
+}
+
+const canToggle = () => {
+  if (!hasRequiredPermissions()) return false
+  if (!isAppReady()) return false
+  if (missingStartNode) return false
+  if (triggerModeDisabled) return false
+  return true
+}
+
+const toggleDisabled = !canToggle()
+```
+
+## Pattern 8: Simplify useMemo/useCallback Dependencies
+
+**Before** (complexity: multiple recalculations):
+
+```typescript
+const payload = useMemo(() => {
+  let parameters: Parameter[] = []
+  let outputParameters: OutputParameter[] = []
+
+  if (!published) {
+    parameters = (inputs || []).map((item) => ({
+      name: item.variable,
+      description: '',
+      form: 'llm',
+      required: item.required,
+      type: item.type,
+    }))
+    outputParameters = (outputs || []).map((item) => ({
+      name: item.variable,
+      description: '',
+      type: item.value_type,
+    }))
+  }
+  else if (detail && detail.tool) {
+    parameters = (inputs || []).map((item) => ({
+      // Complex transformation...
+    }))
+    outputParameters = (outputs || []).map((item) => ({
+      // Complex transformation...
+    }))
+  }
+  
+  return {
+    icon: detail?.icon || icon,
+    label: detail?.label || name,
+    // ...more fields
+  }
+}, [detail, published, workflowAppId, icon, name, description, inputs, outputs])
+```
+
+**After** (complexity: separated concerns):
+
+```typescript
+// Separate transformations
+const useParameterTransform = (inputs: InputVar[], detail?: ToolDetail, published?: boolean) => {
+  return useMemo(() => {
+    if (!published) {
+      return inputs.map(item => ({
+        name: item.variable,
+        description: '',
+        form: 'llm',
+        required: item.required,
+        type: item.type,
+      }))
+    }
+    
+    if (!detail?.tool) return []
+    
+    return inputs.map(item => ({
+      name: item.variable,
+      required: item.required,
+      type: item.type === 'paragraph' ? 'string' : item.type,
+      description: detail.tool.parameters.find(p => p.name === item.variable)?.llm_description || '',
+      form: detail.tool.parameters.find(p => p.name === item.variable)?.form || 'llm',
+    }))
+  }, [inputs, detail, published])
+}
+
+// Component uses hook
+const parameters = useParameterTransform(inputs, detail, published)
+const outputParameters = useOutputTransform(outputs, detail, published)
+
+const payload = useMemo(() => ({
+  icon: detail?.icon || icon,
+  label: detail?.label || name,
+  parameters,
+  outputParameters,
+  // ...
+}), [detail, icon, name, parameters, outputParameters])
+```
+
+## Target Metrics After Refactoring
+
+| Metric | Target |
+|--------|--------|
+| Total Complexity | < 50 |
+| Max Function Complexity | < 30 |
+| Function Length | < 30 lines |
+| Nesting Depth | ≤ 3 levels |
+| Conditional Chains | ≤ 3 conditions |

.claude/skills/component-refactoring/references/component-splitting.md

@@ -0,0 +1,477 @@
+# Component Splitting Patterns
+
+This document provides detailed guidance on splitting large components into smaller, focused components in Dify.
+
+## When to Split Components
+
+Split a component when you identify:
+
+1. **Multiple UI sections** - Distinct visual areas with minimal coupling that can be composed independently
+1. **Conditional rendering blocks** - Large `{condition && <JSX />}` blocks
+1. **Repeated patterns** - Similar UI structures used multiple times
+1. **300+ lines** - Component exceeds manageable size
+1. **Modal clusters** - Multiple modals rendered in one component
+
+## Splitting Strategies
+
+### Strategy 1: Section-Based Splitting
+
+Identify visual sections and extract each as a component.
+
+```typescript
+// ❌ Before: Monolithic component (500+ lines)
+const ConfigurationPage = () => {
+  return (
+    <div>
+      {/* Header Section - 50 lines */}
+      <div className="header">
+        <h1>{t('configuration.title')}</h1>
+        <div className="actions">
+          {isAdvancedMode && <Badge>Advanced</Badge>}
+          <ModelParameterModal ... />
+          <AppPublisher ... />
+        </div>
+      </div>
+      
+      {/* Config Section - 200 lines */}
+      <div className="config">
+        <Config />
+      </div>
+      
+      {/* Debug Section - 150 lines */}
+      <div className="debug">
+        <Debug ... />
+      </div>
+      
+      {/* Modals Section - 100 lines */}
+      {showSelectDataSet && <SelectDataSet ... />}
+      {showHistoryModal && <EditHistoryModal ... />}
+      {showUseGPT4Confirm && <Confirm ... />}
+    </div>
+  )
+}
+
+// ✅ After: Split into focused components
+// configuration/
+//   ├── index.tsx              (orchestration)
+//   ├── configuration-header.tsx
+//   ├── configuration-content.tsx
+//   ├── configuration-debug.tsx
+//   └── configuration-modals.tsx
+
+// configuration-header.tsx
+interface ConfigurationHeaderProps {
+  isAdvancedMode: boolean
+  onPublish: () => void
+}
+
+const ConfigurationHeader: FC<ConfigurationHeaderProps> = ({
+  isAdvancedMode,
+  onPublish,
+}) => {
+  const { t } = useTranslation()
+  
+  return (
+    <div className="header">
+      <h1>{t('configuration.title')}</h1>
+      <div className="actions">
+        {isAdvancedMode && <Badge>Advanced</Badge>}
+        <ModelParameterModal ... />
+        <AppPublisher onPublish={onPublish} />
+      </div>
+    </div>
+  )
+}
+
+// index.tsx (orchestration only)
+const ConfigurationPage = () => {
+  const { modelConfig, setModelConfig } = useModelConfig()
+  const { activeModal, openModal, closeModal } = useModalState()
+  
+  return (
+    <div>
+      <ConfigurationHeader
+        isAdvancedMode={isAdvancedMode}
+        onPublish={handlePublish}
+      />
+      <ConfigurationContent
+        modelConfig={modelConfig}
+        onConfigChange={setModelConfig}
+      />
+      {!isMobile && (
+        <ConfigurationDebug
+          inputs={inputs}
+          onSetting={handleSetting}
+        />
+      )}
+      <ConfigurationModals
+        activeModal={activeModal}
+        onClose={closeModal}
+      />
+    </div>
+  )
+}
+```
+
+### Strategy 2: Conditional Block Extraction
+
+Extract large conditional rendering blocks.
+
+```typescript
+// ❌ Before: Large conditional blocks
+const AppInfo = () => {
+  return (
+    <div>
+      {expand ? (
+        <div className="expanded">
+          {/* 100 lines of expanded view */}
+        </div>
+      ) : (
+        <div className="collapsed">
+          {/* 50 lines of collapsed view */}
+        </div>
+      )}
+    </div>
+  )
+}
+
+// ✅ After: Separate view components
+const AppInfoExpanded: FC<AppInfoViewProps> = ({ appDetail, onAction }) => {
+  return (
+    <div className="expanded">
+      {/* Clean, focused expanded view */}
+    </div>
+  )
+}
+
+const AppInfoCollapsed: FC<AppInfoViewProps> = ({ appDetail, onAction }) => {
+  return (
+    <div className="collapsed">
+      {/* Clean, focused collapsed view */}
+    </div>
+  )
+}
+
+const AppInfo = () => {
+  return (
+    <div>
+      {expand
+        ? <AppInfoExpanded appDetail={appDetail} onAction={handleAction} />
+        : <AppInfoCollapsed appDetail={appDetail} onAction={handleAction} />
+      }
+    </div>
+  )
+}
+```
+
+### Strategy 3: Modal Extraction
+
+Extract modals with their trigger logic.
+
+```typescript
+// ❌ Before: Multiple modals in one component
+const AppInfo = () => {
+  const [showEdit, setShowEdit] = useState(false)
+  const [showDuplicate, setShowDuplicate] = useState(false)
+  const [showDelete, setShowDelete] = useState(false)
+  const [showSwitch, setShowSwitch] = useState(false)
+  
+  const onEdit = async (data) => { /* 20 lines */ }
+  const onDuplicate = async (data) => { /* 20 lines */ }
+  const onDelete = async () => { /* 15 lines */ }
+  
+  return (
+    <div>
+      {/* Main content */}
+      
+      {showEdit && <EditModal onConfirm={onEdit} onClose={() => setShowEdit(false)} />}
+      {showDuplicate && <DuplicateModal onConfirm={onDuplicate} onClose={() => setShowDuplicate(false)} />}
+      {showDelete && <DeleteConfirm onConfirm={onDelete} onClose={() => setShowDelete(false)} />}
+      {showSwitch && <SwitchModal ... />}
+    </div>
+  )
+}
+
+// ✅ After: Modal manager component
+// app-info-modals.tsx
+type ModalType = 'edit' | 'duplicate' | 'delete' | 'switch' | null
+
+interface AppInfoModalsProps {
+  appDetail: AppDetail
+  activeModal: ModalType
+  onClose: () => void
+  onSuccess: () => void
+}
+
+const AppInfoModals: FC<AppInfoModalsProps> = ({
+  appDetail,
+  activeModal,
+  onClose,
+  onSuccess,
+}) => {
+  const handleEdit = async (data) => { /* logic */ }
+  const handleDuplicate = async (data) => { /* logic */ }
+  const handleDelete = async () => { /* logic */ }
+
+  return (
+    <>
+      {activeModal === 'edit' && (
+        <EditModal
+          appDetail={appDetail}
+          onConfirm={handleEdit}
+          onClose={onClose}
+        />
+      )}
+      {activeModal === 'duplicate' && (
+        <DuplicateModal
+          appDetail={appDetail}
+          onConfirm={handleDuplicate}
+          onClose={onClose}
+        />
+      )}
+      {activeModal === 'delete' && (
+        <DeleteConfirm
+          onConfirm={handleDelete}
+          onClose={onClose}
+        />
+      )}
+      {activeModal === 'switch' && (
+        <SwitchModal
+          appDetail={appDetail}
+          onClose={onClose}
+        />
+      )}
+    </>
+  )
+}
+
+// Parent component
+const AppInfo = () => {
+  const { activeModal, openModal, closeModal } = useModalState()
+  
+  return (
+    <div>
+      {/* Main content with openModal triggers */}
+      <Button onClick={() => openModal('edit')}>Edit</Button>
+      
+      <AppInfoModals
+        appDetail={appDetail}
+        activeModal={activeModal}
+        onClose={closeModal}
+        onSuccess={handleSuccess}
+      />
+    </div>
+  )
+}
+```
+
+### Strategy 4: List Item Extraction
+
+Extract repeated item rendering.
+
+```typescript
+// ❌ Before: Inline item rendering
+const OperationsList = () => {
+  return (
+    <div>
+      {operations.map(op => (
+        <div key={op.id} className="operation-item">
+          <span className="icon">{op.icon}</span>
+          <span className="title">{op.title}</span>
+          <span className="description">{op.description}</span>
+          <button onClick={() => op.onClick()}>
+            {op.actionLabel}
+          </button>
+          {op.badge && <Badge>{op.badge}</Badge>}
+          {/* More complex rendering... */}
+        </div>
+      ))}
+    </div>
+  )
+}
+
+// ✅ After: Extracted item component
+interface OperationItemProps {
+  operation: Operation
+  onAction: (id: string) => void
+}
+
+const OperationItem: FC<OperationItemProps> = ({ operation, onAction }) => {
+  return (
+    <div className="operation-item">
+      <span className="icon">{operation.icon}</span>
+      <span className="title">{operation.title}</span>
+      <span className="description">{operation.description}</span>
+      <button onClick={() => onAction(operation.id)}>
+        {operation.actionLabel}
+      </button>
+      {operation.badge && <Badge>{operation.badge}</Badge>}
+    </div>
+  )
+}
+
+const OperationsList = () => {
+  const handleAction = useCallback((id: string) => {
+    const op = operations.find(o => o.id === id)
+    op?.onClick()
+  }, [operations])
+
+  return (
+    <div>
+      {operations.map(op => (
+        <OperationItem
+          key={op.id}
+          operation={op}
+          onAction={handleAction}
+        />
+      ))}
+    </div>
+  )
+}
+```
+
+## Directory Structure Patterns
+
+### Pattern A: Flat Structure (Simple Components)
+
+For components with 2-3 sub-components:
+
+```
+component-name/
+  ├── index.tsx           # Main component
+  ├── sub-component-a.tsx
+  ├── sub-component-b.tsx
+  └── types.ts            # Shared types
+```
+
+### Pattern B: Nested Structure (Complex Components)
+
+For components with many sub-components:
+
+```
+component-name/
+  ├── index.tsx           # Main orchestration
+  ├── types.ts            # Shared types
+  ├── hooks/
+  │   ├── use-feature-a.ts
+  │   └── use-feature-b.ts
+  ├── components/
+  │   ├── header/
+  │   │   └── index.tsx
+  │   ├── content/
+  │   │   └── index.tsx
+  │   └── modals/
+  │       └── index.tsx
+  └── utils/
+      └── helpers.ts
+```
+
+### Pattern C: Feature-Based Structure (Dify Standard)
+
+Following Dify's existing patterns:
+
+```
+configuration/
+  ├── index.tsx           # Main page component
+  ├── base/               # Base/shared components
+  │   ├── feature-panel/
+  │   ├── group-name/
+  │   └── operation-btn/
+  ├── config/             # Config section
+  │   ├── index.tsx
+  │   ├── agent/
+  │   └── automatic/
+  ├── dataset-config/     # Dataset section
+  │   ├── index.tsx
+  │   ├── card-item/
+  │   └── params-config/
+  ├── debug/              # Debug section
+  │   ├── index.tsx
+  │   └── hooks.tsx
+  └── hooks/              # Shared hooks
+      └── use-advanced-prompt-config.ts
+```
+
+## Props Design
+
+### Minimal Props Principle
+
+Pass only what's needed:
+
+```typescript
+// ❌ Bad: Passing entire objects when only some fields needed
+<ConfigHeader appDetail={appDetail} modelConfig={modelConfig} />
+
+// ✅ Good: Destructure to minimum required
+<ConfigHeader
+  appName={appDetail.name}
+  isAdvancedMode={modelConfig.isAdvanced}
+  onPublish={handlePublish}
+/>
+```
+
+### Callback Props Pattern
+
+Use callbacks for child-to-parent communication:
+
+```typescript
+// Parent
+const Parent = () => {
+  const [value, setValue] = useState('')
+  
+  return (
+    <Child
+      value={value}
+      onChange={setValue}
+      onSubmit={handleSubmit}
+    />
+  )
+}
+
+// Child
+interface ChildProps {
+  value: string
+  onChange: (value: string) => void
+  onSubmit: () => void
+}
+
+const Child: FC<ChildProps> = ({ value, onChange, onSubmit }) => {
+  return (
+    <div>
+      <input value={value} onChange={e => onChange(e.target.value)} />
+      <button onClick={onSubmit}>Submit</button>
+    </div>
+  )
+}
+```
+
+### Render Props for Flexibility
+
+When sub-components need parent context:
+
+```typescript
+interface ListProps<T> {
+  items: T[]
+  renderItem: (item: T, index: number) => React.ReactNode
+  renderEmpty?: () => React.ReactNode
+}
+
+function List<T>({ items, renderItem, renderEmpty }: ListProps<T>) {
+  if (items.length === 0 && renderEmpty) {
+    return <>{renderEmpty()}</>
+  }
+  
+  return (
+    <div>
+      {items.map((item, index) => renderItem(item, index))}
+    </div>
+  )
+}
+
+// Usage
+<List
+  items={operations}
+  renderItem={(op, i) => <OperationItem key={i} operation={op} />}
+  renderEmpty={() => <EmptyState message="No operations" />}
+/>
+```

.claude/skills/component-refactoring/references/hook-extraction.md

@@ -0,0 +1,317 @@
+# Hook Extraction Patterns
+
+This document provides detailed guidance on extracting custom hooks from complex components in Dify.
+
+## When to Extract Hooks
+
+Extract a custom hook when you identify:
+
+1. **Coupled state groups** - Multiple `useState` hooks that are always used together
+1. **Complex effects** - `useEffect` with multiple dependencies or cleanup logic
+1. **Business logic** - Data transformations, validations, or calculations
+1. **Reusable patterns** - Logic that appears in multiple components
+
+## Extraction Process
+
+### Step 1: Identify State Groups
+
+Look for state variables that are logically related:
+
+```typescript
+// ❌ These belong together - extract to hook
+const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+const [completionParams, setCompletionParams] = useState<FormValue>({})
+const [modelModeType, setModelModeType] = useState<ModelModeType>(...)
+
+// These are model-related state that should be in useModelConfig()
+```
+
+### Step 2: Identify Related Effects
+
+Find effects that modify the grouped state:
+
+```typescript
+// ❌ These effects belong with the state above
+useEffect(() => {
+  if (hasFetchedDetail && !modelModeType) {
+    const mode = currModel?.model_properties.mode
+    if (mode) {
+      const newModelConfig = produce(modelConfig, (draft) => {
+        draft.mode = mode
+      })
+      setModelConfig(newModelConfig)
+    }
+  }
+}, [textGenerationModelList, hasFetchedDetail, modelModeType, currModel])
+```
+
+### Step 3: Create the Hook
+
+```typescript
+// hooks/use-model-config.ts
+import type { FormValue } from '@/app/components/header/account-setting/model-provider-page/declarations'
+import type { ModelConfig } from '@/models/debug'
+import { produce } from 'immer'
+import { useEffect, useState } from 'react'
+import { ModelModeType } from '@/types/app'
+
+interface UseModelConfigParams {
+  initialConfig?: Partial<ModelConfig>
+  currModel?: { model_properties?: { mode?: ModelModeType } }
+  hasFetchedDetail: boolean
+}
+
+interface UseModelConfigReturn {
+  modelConfig: ModelConfig
+  setModelConfig: (config: ModelConfig) => void
+  completionParams: FormValue
+  setCompletionParams: (params: FormValue) => void
+  modelModeType: ModelModeType
+}
+
+export const useModelConfig = ({
+  initialConfig,
+  currModel,
+  hasFetchedDetail,
+}: UseModelConfigParams): UseModelConfigReturn => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>({
+    provider: 'langgenius/openai/openai',
+    model_id: 'gpt-3.5-turbo',
+    mode: ModelModeType.unset,
+    // ... default values
+    ...initialConfig,
+  })
+  
+  const [completionParams, setCompletionParams] = useState<FormValue>({})
+  
+  const modelModeType = modelConfig.mode
+
+  // Fill old app data missing model mode
+  useEffect(() => {
+    if (hasFetchedDetail && !modelModeType) {
+      const mode = currModel?.model_properties?.mode
+      if (mode) {
+        setModelConfig(produce(modelConfig, (draft) => {
+          draft.mode = mode
+        }))
+      }
+    }
+  }, [hasFetchedDetail, modelModeType, currModel])
+
+  return {
+    modelConfig,
+    setModelConfig,
+    completionParams,
+    setCompletionParams,
+    modelModeType,
+  }
+}
+```
+
+### Step 4: Update Component
+
+```typescript
+// Before: 50+ lines of state management
+const Configuration: FC = () => {
+  const [modelConfig, setModelConfig] = useState<ModelConfig>(...)
+  // ... lots of related state and effects
+}
+
+// After: Clean component
+const Configuration: FC = () => {
+  const {
+    modelConfig,
+    setModelConfig,
+    completionParams,
+    setCompletionParams,
+    modelModeType,
+  } = useModelConfig({
+    currModel,
+    hasFetchedDetail,
+  })
+  
+  // Component now focuses on UI
+}
+```
+
+## Naming Conventions
+
+### Hook Names
+
+- Use `use` prefix: `useModelConfig`, `useDatasetConfig`
+- Be specific: `useAdvancedPromptConfig` not `usePrompt`
+- Include domain: `useWorkflowVariables`, `useMCPServer`
+
+### File Names
+
+- Kebab-case: `use-model-config.ts`
+- Place in `hooks/` subdirectory when multiple hooks exist
+- Place alongside component for single-use hooks
+
+### Return Type Names
+
+- Suffix with `Return`: `UseModelConfigReturn`
+- Suffix params with `Params`: `UseModelConfigParams`
+
+## Common Hook Patterns in Dify
+
+### 1. Data Fetching Hook (React Query)
+
+```typescript
+// Pattern: Use @tanstack/react-query for data fetching
+import { useQuery, useQueryClient } from '@tanstack/react-query'
+import { get } from '@/service/base'
+import { useInvalid } from '@/service/use-base'
+
+const NAME_SPACE = 'appConfig'
+
+// Query keys for cache management
+export const appConfigQueryKeys = {
+  detail: (appId: string) => [NAME_SPACE, 'detail', appId] as const,
+}
+
+// Main data hook
+export const useAppConfig = (appId: string) => {
+  return useQuery({
+    enabled: !!appId,
+    queryKey: appConfigQueryKeys.detail(appId),
+    queryFn: () => get<AppDetailResponse>(`/apps/${appId}`),
+    select: data => data?.model_config || null,
+  })
+}
+
+// Invalidation hook for refreshing data
+export const useInvalidAppConfig = () => {
+  return useInvalid([NAME_SPACE])
+}
+
+// Usage in component
+const Component = () => {
+  const { data: config, isLoading, error, refetch } = useAppConfig(appId)
+  const invalidAppConfig = useInvalidAppConfig()
+  
+  const handleRefresh = () => {
+    invalidAppConfig() // Invalidates cache and triggers refetch
+  }
+  
+  return <div>...</div>
+}
+```
+
+### 2. Form State Hook
+
+```typescript
+// Pattern: Form state + validation + submission
+export const useConfigForm = (initialValues: ConfigFormValues) => {
+  const [values, setValues] = useState(initialValues)
+  const [errors, setErrors] = useState<Record<string, string>>({})
+  const [isSubmitting, setIsSubmitting] = useState(false)
+
+  const validate = useCallback(() => {
+    const newErrors: Record<string, string> = {}
+    if (!values.name) newErrors.name = 'Name is required'
+    setErrors(newErrors)
+    return Object.keys(newErrors).length === 0
+  }, [values])
+
+  const handleChange = useCallback((field: string, value: any) => {
+    setValues(prev => ({ ...prev, [field]: value }))
+  }, [])
+
+  const handleSubmit = useCallback(async (onSubmit: (values: ConfigFormValues) => Promise<void>) => {
+    if (!validate()) return
+    setIsSubmitting(true)
+    try {
+      await onSubmit(values)
+    } finally {
+      setIsSubmitting(false)
+    }
+  }, [values, validate])
+
+  return { values, errors, isSubmitting, handleChange, handleSubmit }
+}
+```
+
+### 3. Modal State Hook
+
+```typescript
+// Pattern: Multiple modal management
+type ModalType = 'edit' | 'delete' | 'duplicate' | null
+
+export const useModalState = () => {
+  const [activeModal, setActiveModal] = useState<ModalType>(null)
+  const [modalData, setModalData] = useState<any>(null)
+
+  const openModal = useCallback((type: ModalType, data?: any) => {
+    setActiveModal(type)
+    setModalData(data)
+  }, [])
+
+  const closeModal = useCallback(() => {
+    setActiveModal(null)
+    setModalData(null)
+  }, [])
+
+  return {
+    activeModal,
+    modalData,
+    openModal,
+    closeModal,
+    isOpen: useCallback((type: ModalType) => activeModal === type, [activeModal]),
+  }
+}
+```
+
+### 4. Toggle/Boolean Hook
+
+```typescript
+// Pattern: Boolean state with convenience methods
+export const useToggle = (initialValue = false) => {
+  const [value, setValue] = useState(initialValue)
+
+  const toggle = useCallback(() => setValue(v => !v), [])
+  const setTrue = useCallback(() => setValue(true), [])
+  const setFalse = useCallback(() => setValue(false), [])
+
+  return [value, { toggle, setTrue, setFalse, set: setValue }] as const
+}
+
+// Usage
+const [isExpanded, { toggle, setTrue: expand, setFalse: collapse }] = useToggle()
+```
+
+## Testing Extracted Hooks
+
+After extraction, test hooks in isolation:
+
+```typescript
+// use-model-config.spec.ts
+import { renderHook, act } from '@testing-library/react'
+import { useModelConfig } from './use-model-config'
+
+describe('useModelConfig', () => {
+  it('should initialize with default values', () => {
+    const { result } = renderHook(() => useModelConfig({
+      hasFetchedDetail: false,
+    }))
+
+    expect(result.current.modelConfig.provider).toBe('langgenius/openai/openai')
+    expect(result.current.modelModeType).toBe(ModelModeType.unset)
+  })
+
+  it('should update model config', () => {
+    const { result } = renderHook(() => useModelConfig({
+      hasFetchedDetail: true,
+    }))
+
+    act(() => {
+      result.current.setModelConfig({
+        ...result.current.modelConfig,
+        model_id: 'gpt-4',
+      })
+    })
+
+    expect(result.current.modelConfig.model_id).toBe('gpt-4')
+  })
+})
+```

.claude/skills/frontend-testing/SKILL.md

@@ -318,5 +318,5 @@ For more detailed information, refer to:
 
 - `web/vitest.config.ts` - Vitest configuration
 - `web/vitest.setup.ts` - Test environment setup
-- `web/testing/analyze-component.js` - Component analysis tool
+- `web/scripts/analyze-component.js` - Component analysis tool
 - Modules are not mocked automatically. Global mocks live in `web/vitest.setup.ts` (for example `react-i18next`, `next/image`); mock other modules like `ky` or `mime` locally in test files.

.claude/skills/frontend-testing/assets/component-test.template.tsx

@@ -28,17 +28,14 @@ import userEvent from '@testing-library/user-event'
 
 // i18n (automatically mocked)
 // WHY: Global mock in web/vitest.setup.ts is auto-loaded by Vitest setup
-// No explicit mock needed - it returns translation keys as-is
+// The global mock provides: useTranslation, Trans, useMixedTranslation, useGetLanguage
+// No explicit mock needed for most tests
+//
 // Override only if custom translations are required:
-// vi.mock('react-i18next', () => ({
-//   useTranslation: () => ({
-//     t: (key: string) => {
-//       const customTranslations: Record<string, string> = {
-//         'my.custom.key': 'Custom Translation',
-//       }
-//       return customTranslations[key] || key
-//     },
-//   }),
+// import { createReactI18nextMock } from '@/test/i18n-mock'
+// vi.mock('react-i18next', () => createReactI18nextMock({
+//   'my.custom.key': 'Custom Translation',
+//   'button.save': 'Save',
 // }))
 
 // Router (if component uses useRouter, usePathname, useSearchParams)

.claude/skills/frontend-testing/references/mocking.md

@@ -52,23 +52,29 @@ Modules are not mocked automatically. Use `vi.mock` in test files, or add global
 ### 1. i18n (Auto-loaded via Global Mock)
 
 A global mock is defined in `web/vitest.setup.ts` and is auto-loaded by Vitest setup.
-**No explicit mock needed** for most tests - it returns translation keys as-is.
 
-For tests requiring custom translations, override the mock:
+The global mock provides:
+
+- `useTranslation` - returns translation keys with namespace prefix
+- `Trans` component - renders i18nKey and components
+- `useMixedTranslation` (from `@/app/components/plugins/marketplace/hooks`)
+- `useGetLanguage` (from `@/context/i18n`) - returns `'en-US'`
+
+**Default behavior**: Most tests should use the global mock (no local override needed).
+
+**For custom translations**: Use the helper function from `@/test/i18n-mock`:
 
 ```typescript
-vi.mock('react-i18next', () => ({
-  useTranslation: () => ({
-    t: (key: string) => {
-      const translations: Record<string, string> = {
-        'my.custom.key': 'Custom translation',
-      }
-      return translations[key] || key
-    },
-  }),
+import { createReactI18nextMock } from '@/test/i18n-mock'
+
+vi.mock('react-i18next', () => createReactI18nextMock({
+  'my.custom.key': 'Custom translation',
+  'button.save': 'Save',
 }))
 ```
 
+**Avoid**: Manually defining `useTranslation` mocks that just return the key - the global mock already does this.
+
 ### 2. Next.js Router
 
 ```typescript

.github/workflows/api-tests.yml

@@ -22,12 +22,12 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: ${{ matrix.python-version }}
@@ -57,7 +57,7 @@ jobs:
         run: sh .github/workflows/expose_service_ports.sh
 
       - name: Set up Sandbox
-        uses: hoverkraft-tech/compose-action@v2.0.2
+        uses: hoverkraft-tech/compose-action@v2
         with:
           compose-file: |
             docker/docker-compose.middleware.yaml

.github/workflows/autofix.yml

@@ -12,7 +12,7 @@ jobs:
     if: github.repository == 'langgenius/dify'
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
 
       - name: Check Docker Compose inputs
         id: docker-compose-changes
@@ -27,7 +27,7 @@ jobs:
         with:
           python-version: "3.11"
 
-      - uses: astral-sh/setup-uv@v6
+      - uses: astral-sh/setup-uv@v7
 
       - name: Generate Docker Compose
         if: steps.docker-compose-changes.outputs.any_changed == 'true'

.github/workflows/build-push.yml

@@ -90,7 +90,7 @@ jobs:
           touch "/tmp/digests/${sanitized_digest}"
 
       - name: Upload digest
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v6
         with:
           name: digests-${{ matrix.context }}-${{ env.PLATFORM_PAIR }}
           path: /tmp/digests/*

.github/workflows/db-migration-test.yml

@@ -13,13 +13,13 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: "3.12"
@@ -63,13 +63,13 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: "3.12"

.github/workflows/main-ci.yml

@@ -27,7 +27,7 @@ jobs:
       vdb-changed: ${{ steps.changes.outputs.vdb }}
       migration-changed: ${{ steps.changes.outputs.migration }}
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
       - uses: dorny/paths-filter@v3
         id: changes
         with:
@@ -38,6 +38,7 @@ jobs:
               - '.github/workflows/api-tests.yml'
             web:
               - 'web/**'
+              - '.github/workflows/web-tests.yml'
             vdb:
               - 'api/core/rag/datasource/**'
               - 'docker/**'

.github/workflows/style.yml

@@ -19,13 +19,13 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Check changed files
         id: changed-files
-        uses: tj-actions/changed-files@v46
+        uses: tj-actions/changed-files@v47
         with:
           files: |
             api/**
@@ -33,7 +33,7 @@ jobs:
 
       - name: Setup UV and Python
         if: steps.changed-files.outputs.any_changed == 'true'
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: false
           python-version: "3.12"
@@ -68,15 +68,17 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Check changed files
         id: changed-files
-        uses: tj-actions/changed-files@v46
+        uses: tj-actions/changed-files@v47
         with:
-          files: web/**
+          files: |
+            web/**
+            .github/workflows/style.yml
 
       - name: Install pnpm
         uses: pnpm/action-setup@v4
@@ -85,7 +87,7 @@ jobs:
           run_install: false
 
       - name: Setup NodeJS
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         if: steps.changed-files.outputs.any_changed == 'true'
         with:
           node-version: 22
@@ -108,20 +110,30 @@ jobs:
         working-directory: ./web
         run: pnpm run type-check:tsgo
 
+      - name: Web dead code check
+        if: steps.changed-files.outputs.any_changed == 'true'
+        working-directory: ./web
+        run: pnpm run knip
+
+      - name: Web build check
+        if: steps.changed-files.outputs.any_changed == 'true'
+        working-directory: ./web
+        run: pnpm run build
+
   superlinter:
     name: SuperLinter
     runs-on: ubuntu-latest
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           fetch-depth: 0
           persist-credentials: false
 
       - name: Check changed files
         id: changed-files
-        uses: tj-actions/changed-files@v46
+        uses: tj-actions/changed-files@v47
         with:
           files: |
             **.sh

.github/workflows/tool-test-sdks.yaml

@@ -25,12 +25,12 @@ jobs:
         working-directory: sdks/nodejs-client
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Use Node.js ${{ matrix.node-version }}
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         with:
           node-version: ${{ matrix.node-version }}
           cache: ''

.github/workflows/translate-i18n-base-on-english.yml

@@ -4,7 +4,7 @@ on:
   push:
     branches: [main]
     paths:
-      - 'web/i18n/en-US/*.ts'
+      - 'web/i18n/en-US/*.json'
 
 permissions:
   contents: write
@@ -18,7 +18,7 @@ jobs:
       run:
         working-directory: web
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
         with:
           fetch-depth: 0
           token: ${{ secrets.GITHUB_TOKEN }}
@@ -28,13 +28,13 @@ jobs:
         run: |
           git fetch origin "${{ github.event.before }}" || true
           git fetch origin "${{ github.sha }}" || true
-          changed_files=$(git diff --name-only "${{ github.event.before }}" "${{ github.sha }}" -- 'i18n/en-US/*.ts')
+          changed_files=$(git diff --name-only "${{ github.event.before }}" "${{ github.sha }}" -- 'i18n/en-US/*.json')
           echo "Changed files: $changed_files"
           if [ -n "$changed_files" ]; then
             echo "FILES_CHANGED=true" >> $GITHUB_ENV
             file_args=""
             for file in $changed_files; do
-              filename=$(basename "$file" .ts)
+              filename=$(basename "$file" .json)
               file_args="$file_args --file $filename"
             done
             echo "FILE_ARGS=$file_args" >> $GITHUB_ENV
@@ -51,7 +51,7 @@ jobs:
 
       - name: Set up Node.js
         if: env.FILES_CHANGED == 'true'
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         with:
           node-version: 'lts/*'
           cache: pnpm
@@ -65,7 +65,7 @@ jobs:
       - name: Generate i18n translations
         if: env.FILES_CHANGED == 'true'
         working-directory: ./web
-        run: pnpm run auto-gen-i18n ${{ env.FILE_ARGS }}
+        run: pnpm run i18n:gen ${{ env.FILE_ARGS }}
 
       - name: Create Pull Request
         if: env.FILES_CHANGED == 'true'

.github/workflows/vdb-tests.yml

@@ -19,19 +19,19 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
       - name: Free Disk Space
-        uses: endersonmenezes/free-disk-space@v2
+        uses: endersonmenezes/free-disk-space@v3
         with:
           remove_dotnet: true
           remove_haskell: true
           remove_tool_cache: true
 
       - name: Setup UV and Python
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
         with:
           enable-cache: true
           python-version: ${{ matrix.python-version }}

.github/workflows/web-tests.yml

@@ -18,7 +18,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
 
@@ -29,7 +29,7 @@ jobs:
           run_install: false
 
       - name: Setup Node.js
-        uses: actions/setup-node@v4
+        uses: actions/setup-node@v6
         with:
           node-version: 22
           cache: pnpm
@@ -360,7 +360,7 @@ jobs:
 
       - name: Upload Coverage Artifact
         if: steps.coverage-summary.outputs.has_coverage == 'true'
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v6
         with:
           name: web-coverage-report
           path: web/coverage

.mcp.json

@@ -1,34 +0,0 @@
-{
-    "mcpServers": {
-      "context7": {
-        "type": "http",
-        "url": "https://mcp.context7.com/mcp"
-      },
-      "sequential-thinking": {
-        "type": "stdio",
-        "command": "npx",
-        "args": ["-y", "@modelcontextprotocol/server-sequential-thinking"],
-        "env": {}
-      },
-      "github": {
-        "type": "stdio",
-        "command": "npx",
-        "args": ["-y", "@modelcontextprotocol/server-github"],
-        "env": {
-          "GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_PERSONAL_ACCESS_TOKEN}"
-        }
-      },
-      "fetch": {
-        "type": "stdio",
-        "command": "uvx",
-        "args": ["mcp-server-fetch"],
-        "env": {}
-      },
-      "playwright": {
-        "type": "stdio",
-        "command": "npx",
-        "args": ["-y", "@playwright/mcp@latest"],
-        "env": {}
-      }
-    }
-  }

api/.env.example

@@ -101,6 +101,15 @@ S3_ACCESS_KEY=your-access-key
 S3_SECRET_KEY=your-secret-key
 S3_REGION=your-region
 
+# Workflow run and Conversation archive storage (S3-compatible)
+ARCHIVE_STORAGE_ENABLED=false
+ARCHIVE_STORAGE_ENDPOINT=
+ARCHIVE_STORAGE_ARCHIVE_BUCKET=
+ARCHIVE_STORAGE_EXPORT_BUCKET=
+ARCHIVE_STORAGE_ACCESS_KEY=
+ARCHIVE_STORAGE_SECRET_KEY=
+ARCHIVE_STORAGE_REGION=auto
+
 # Azure Blob Storage configuration
 AZURE_BLOB_ACCOUNT_NAME=your-account-name
 AZURE_BLOB_ACCOUNT_KEY=your-account-key
@@ -128,6 +137,7 @@ TENCENT_COS_SECRET_KEY=your-secret-key
 TENCENT_COS_SECRET_ID=your-secret-id
 TENCENT_COS_REGION=your-region
 TENCENT_COS_SCHEME=your-scheme
+TENCENT_COS_CUSTOM_DOMAIN=your-custom-domain
 
 # Huawei OBS Storage Configuration
 HUAWEI_OBS_BUCKET_NAME=your-bucket-name
@@ -492,6 +502,8 @@ LOG_FILE_BACKUP_COUNT=5
 LOG_DATEFORMAT=%Y-%m-%d %H:%M:%S
 # Log Timezone
 LOG_TZ=UTC
+# Log output format: text or json
+LOG_OUTPUT_FORMAT=text
 # Log format
 LOG_FORMAT=%(asctime)s,%(msecs)d %(levelname)-2s [%(filename)s:%(lineno)d] %(req_id)s %(message)s
 

api/.ruff.toml

@@ -1,4 +1,8 @@
-exclude = ["migrations/*"]
+exclude = [
+    "migrations/*",
+    ".git",
+    ".git/**",
+]
 line-length = 120
 
 [format]

api/app_factory.py

@@ -2,9 +2,11 @@ import logging
 import time
 
 from opentelemetry.trace import get_current_span
+from opentelemetry.trace.span import INVALID_SPAN_ID, INVALID_TRACE_ID
 
 from configs import dify_config
 from contexts.wrapper import RecyclableContextVar
+from core.logging.context import init_request_context
 from dify_app import DifyApp
 
 logger = logging.getLogger(__name__)
@@ -25,28 +27,35 @@ def create_flask_app_with_configs() -> DifyApp:
     # add before request hook
     @dify_app.before_request
     def before_request():
-        # add an unique identifier to each request
+        # Initialize logging context for this request
+        init_request_context()
         RecyclableContextVar.increment_thread_recycles()
 
-    # add after request hook for injecting X-Trace-Id header from OpenTelemetry span context
+    # add after request hook for injecting trace headers from OpenTelemetry span context
+    # Only adds headers when OTEL is enabled and has valid context
     @dify_app.after_request
-    def add_trace_id_header(response):
+    def add_trace_headers(response):
         try:
             span = get_current_span()
             ctx = span.get_span_context() if span else None
-            if ctx and ctx.is_valid:
-                trace_id_hex = format(ctx.trace_id, "032x")
-                # Avoid duplicates if some middleware added it
-                if "X-Trace-Id" not in response.headers:
-                    response.headers["X-Trace-Id"] = trace_id_hex
+
+            if not ctx or not ctx.is_valid:
+                return response
+
+            # Inject trace headers from OTEL context
+            if ctx.trace_id != INVALID_TRACE_ID and "X-Trace-Id" not in response.headers:
+                response.headers["X-Trace-Id"] = format(ctx.trace_id, "032x")
+            if ctx.span_id != INVALID_SPAN_ID and "X-Span-Id" not in response.headers:
+                response.headers["X-Span-Id"] = format(ctx.span_id, "016x")
+
         except Exception:
             # Never break the response due to tracing header injection
-            logger.warning("Failed to add trace ID to response header", exc_info=True)
+            logger.warning("Failed to add trace headers to response", exc_info=True)
         return response
 
     # Capture the decorator's return value to avoid pyright reportUnusedFunction
     _ = before_request
-    _ = add_trace_id_header
+    _ = add_trace_headers
 
     return dify_app
 

api/configs/extra/__init__.py

@@ -1,9 +1,11 @@
+from configs.extra.archive_config import ArchiveStorageConfig
 from configs.extra.notion_config import NotionConfig
 from configs.extra.sentry_config import SentryConfig
 
 
 class ExtraServiceConfig(
     # place the configs in alphabet order
+    ArchiveStorageConfig,
     NotionConfig,
     SentryConfig,
 ):

api/configs/extra/archive_config.py

@@ -0,0 +1,43 @@
+from pydantic import Field
+from pydantic_settings import BaseSettings
+
+
+class ArchiveStorageConfig(BaseSettings):
+    """
+    Configuration settings for workflow run logs archiving storage.
+    """
+
+    ARCHIVE_STORAGE_ENABLED: bool = Field(
+        description="Enable workflow run logs archiving to S3-compatible storage",
+        default=False,
+    )
+
+    ARCHIVE_STORAGE_ENDPOINT: str | None = Field(
+        description="URL of the S3-compatible storage endpoint (e.g., 'https://storage.example.com')",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_ARCHIVE_BUCKET: str | None = Field(
+        description="Name of the bucket to store archived workflow logs",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_EXPORT_BUCKET: str | None = Field(
+        description="Name of the bucket to store exported workflow runs",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_ACCESS_KEY: str | None = Field(
+        description="Access key ID for authenticating with storage",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_SECRET_KEY: str | None = Field(
+        description="Secret access key for authenticating with storage",
+        default=None,
+    )
+
+    ARCHIVE_STORAGE_REGION: str = Field(
+        description="Region for storage (use 'auto' if the provider supports it)",
+        default="auto",
+    )

api/configs/feature/__init__.py

@@ -587,6 +587,11 @@ class LoggingConfig(BaseSettings):
         default="INFO",
     )
 
+    LOG_OUTPUT_FORMAT: Literal["text", "json"] = Field(
+        description="Log output format: 'text' for human-readable, 'json' for structured JSON logs.",
+        default="text",
+    )
+
     LOG_FILE: str | None = Field(
         description="File path for log output.",
         default=None,

api/configs/middleware/storage/tencent_cos_storage_config.py

@@ -31,3 +31,8 @@ class TencentCloudCOSStorageConfig(BaseSettings):
         description="Protocol scheme for COS requests: 'https' (recommended) or 'http'",
         default=None,
     )
+
+    TENCENT_COS_CUSTOM_DOMAIN: str | None = Field(
+        description="Tencent Cloud COS custom domain setting",
+        default=None,
+    )

api/controllers/common/fields.py

@@ -1,62 +1,59 @@
-from flask_restx import Api, Namespace, fields
+from __future__ import annotations
 
-from libs.helper import AppIconUrlField
+from typing import Any, TypeAlias
 
-parameters__system_parameters = {
-    "image_file_size_limit": fields.Integer,
-    "video_file_size_limit": fields.Integer,
-    "audio_file_size_limit": fields.Integer,
-    "file_size_limit": fields.Integer,
-    "workflow_file_upload_limit": fields.Integer,
-}
+from pydantic import BaseModel, ConfigDict, computed_field
+
+from core.file import helpers as file_helpers
+from models.model import IconType
+
+JSONValue: TypeAlias = str | int | float | bool | None | dict[str, Any] | list[Any]
+JSONObject: TypeAlias = dict[str, Any]
 
 
-def build_system_parameters_model(api_or_ns: Api | Namespace):
-    """Build the system parameters model for the API or Namespace."""
-    return api_or_ns.model("SystemParameters", parameters__system_parameters)
+class SystemParameters(BaseModel):
+    image_file_size_limit: int
+    video_file_size_limit: int
+    audio_file_size_limit: int
+    file_size_limit: int
+    workflow_file_upload_limit: int
 
 
-parameters_fields = {
-    "opening_statement": fields.String,
-    "suggested_questions": fields.Raw,
-    "suggested_questions_after_answer": fields.Raw,
-    "speech_to_text": fields.Raw,
-    "text_to_speech": fields.Raw,
-    "retriever_resource": fields.Raw,
-    "annotation_reply": fields.Raw,
-    "more_like_this": fields.Raw,
-    "user_input_form": fields.Raw,
-    "sensitive_word_avoidance": fields.Raw,
-    "file_upload": fields.Raw,
-    "system_parameters": fields.Nested(parameters__system_parameters),
-}
+class Parameters(BaseModel):
+    opening_statement: str | None = None
+    suggested_questions: list[str]
+    suggested_questions_after_answer: JSONObject
+    speech_to_text: JSONObject
+    text_to_speech: JSONObject
+    retriever_resource: JSONObject
+    annotation_reply: JSONObject
+    more_like_this: JSONObject
+    user_input_form: list[JSONObject]
+    sensitive_word_avoidance: JSONObject
+    file_upload: JSONObject
+    system_parameters: SystemParameters
 
 
-def build_parameters_model(api_or_ns: Api | Namespace):
-    """Build the parameters model for the API or Namespace."""
-    copied_fields = parameters_fields.copy()
-    copied_fields["system_parameters"] = fields.Nested(build_system_parameters_model(api_or_ns))
-    return api_or_ns.model("Parameters", copied_fields)
+class Site(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
 
+    title: str
+    chat_color_theme: str | None = None
+    chat_color_theme_inverted: bool
+    icon_type: str | None = None
+    icon: str | None = None
+    icon_background: str | None = None
+    description: str | None = None
+    copyright: str | None = None
+    privacy_policy: str | None = None
+    custom_disclaimer: str | None = None
+    default_language: str
+    show_workflow_steps: bool
+    use_icon_as_answer_icon: bool
 
-site_fields = {
-    "title": fields.String,
-    "chat_color_theme": fields.String,
-    "chat_color_theme_inverted": fields.Boolean,
-    "icon_type": fields.String,
-    "icon": fields.String,
-    "icon_background": fields.String,
-    "icon_url": AppIconUrlField,
-    "description": fields.String,
-    "copyright": fields.String,
-    "privacy_policy": fields.String,
-    "custom_disclaimer": fields.String,
-    "default_language": fields.String,
-    "show_workflow_steps": fields.Boolean,
-    "use_icon_as_answer_icon": fields.Boolean,
-}
-
-
-def build_site_model(api_or_ns: Api | Namespace):
-    """Build the site model for the API or Namespace."""
-    return api_or_ns.model("Site", site_fields)
+    @computed_field(return_type=str | None)  # type: ignore
+    @property
+    def icon_url(self) -> str | None:
+        if self.icon and self.icon_type == IconType.IMAGE:
+            return file_helpers.get_signed_file_url(self.icon)
+        return None

api/controllers/console/app/app.py

@@ -1,3 +1,4 @@
+import re
 import uuid
 from typing import Literal
 
@@ -73,6 +74,48 @@ class AppListQuery(BaseModel):
             raise ValueError("Invalid UUID format in tag_ids.") from exc
 
 
+# XSS prevention: patterns that could lead to XSS attacks
+# Includes: script tags, iframe tags, javascript: protocol, SVG with onload, etc.
+_XSS_PATTERNS = [
+    r"<script[^>]*>.*?</script>",  # Script tags
+    r"<iframe\b[^>]*?(?:/>|>.*?</iframe>)",  # Iframe tags (including self-closing)
+    r"javascript:",  # JavaScript protocol
+    r"<svg[^>]*?\s+onload\s*=[^>]*>",  # SVG with onload handler (attribute-aware, flexible whitespace)
+    r"<.*?on\s*\w+\s*=",  # Event handlers like onclick, onerror, etc.
+    r"<object\b[^>]*(?:\s*/>|>.*?</object\s*>)",  # Object tags (opening tag)
+    r"<embed[^>]*>",  # Embed tags (self-closing)
+    r"<link[^>]*>",  # Link tags with javascript
+]
+
+
+def _validate_xss_safe(value: str | None, field_name: str = "Field") -> str | None:
+    """
+    Validate that a string value doesn't contain potential XSS payloads.
+
+    Args:
+        value: The string value to validate
+        field_name: Name of the field for error messages
+
+    Returns:
+        The original value if safe
+
+    Raises:
+        ValueError: If the value contains XSS patterns
+    """
+    if value is None:
+        return None
+
+    value_lower = value.lower()
+    for pattern in _XSS_PATTERNS:
+        if re.search(pattern, value_lower, re.DOTALL | re.IGNORECASE):
+            raise ValueError(
+                f"{field_name} contains invalid characters or patterns. "
+                "HTML tags, JavaScript, and other potentially dangerous content are not allowed."
+            )
+
+    return value
+
+
 class CreateAppPayload(BaseModel):
     name: str = Field(..., min_length=1, description="App name")
     description: str | None = Field(default=None, description="App description (max 400 chars)", max_length=400)
@@ -81,6 +124,11 @@ class CreateAppPayload(BaseModel):
     icon: str | None = Field(default=None, description="Icon")
     icon_background: str | None = Field(default=None, description="Icon background color")
 
+    @field_validator("name", "description", mode="before")
+    @classmethod
+    def validate_xss_safe(cls, value: str | None, info) -> str | None:
+        return _validate_xss_safe(value, info.field_name)
+
 
 class UpdateAppPayload(BaseModel):
     name: str = Field(..., min_length=1, description="App name")
@@ -91,6 +139,11 @@ class UpdateAppPayload(BaseModel):
     use_icon_as_answer_icon: bool | None = Field(default=None, description="Use icon as answer icon")
     max_active_requests: int | None = Field(default=None, description="Maximum active requests")
 
+    @field_validator("name", "description", mode="before")
+    @classmethod
+    def validate_xss_safe(cls, value: str | None, info) -> str | None:
+        return _validate_xss_safe(value, info.field_name)
+
 
 class CopyAppPayload(BaseModel):
     name: str | None = Field(default=None, description="Name for the copied app")
@@ -99,6 +152,11 @@ class CopyAppPayload(BaseModel):
     icon: str | None = Field(default=None, description="Icon")
     icon_background: str | None = Field(default=None, description="Icon background color")
 
+    @field_validator("name", "description", mode="before")
+    @classmethod
+    def validate_xss_safe(cls, value: str | None, info) -> str | None:
+        return _validate_xss_safe(value, info.field_name)
+
 
 class AppExportQuery(BaseModel):
     include_secret: bool = Field(default=False, description="Include secrets in export")

api/controllers/console/auth/oauth.py

@@ -124,7 +124,7 @@ class OAuthCallback(Resource):
             return redirect(f"{dify_config.CONSOLE_WEB_URL}/signin/invite-settings?invite_token={invite_token}")
 
         try:
-            account = _generate_account(provider, user_info)
+            account, oauth_new_user = _generate_account(provider, user_info)
         except AccountNotFoundError:
             return redirect(f"{dify_config.CONSOLE_WEB_URL}/signin?message=Account not found.")
         except (WorkSpaceNotFoundError, WorkSpaceNotAllowedCreateError):
@@ -159,7 +159,10 @@ class OAuthCallback(Resource):
             ip_address=extract_remote_ip(request),
         )
 
-        response = redirect(f"{dify_config.CONSOLE_WEB_URL}")
+        base_url = dify_config.CONSOLE_WEB_URL
+        query_char = "&" if "?" in base_url else "?"
+        target_url = f"{base_url}{query_char}oauth_new_user={str(oauth_new_user).lower()}"
+        response = redirect(target_url)
 
         set_access_token_to_cookie(request, response, token_pair.access_token)
         set_refresh_token_to_cookie(request, response, token_pair.refresh_token)
@@ -177,9 +180,10 @@ def _get_account_by_openid_or_email(provider: str, user_info: OAuthUserInfo) ->
     return account
 
 
-def _generate_account(provider: str, user_info: OAuthUserInfo):
+def _generate_account(provider: str, user_info: OAuthUserInfo) -> tuple[Account, bool]:
     # Get account by openid or email.
     account = _get_account_by_openid_or_email(provider, user_info)
+    oauth_new_user = False
 
     if account:
         tenants = TenantService.get_join_tenants(account)
@@ -193,6 +197,7 @@ def _generate_account(provider: str, user_info: OAuthUserInfo):
                 tenant_was_created.send(new_tenant)
 
     if not account:
+        oauth_new_user = True
         if not FeatureService.get_system_features().is_allow_register:
             if dify_config.BILLING_ENABLED and BillingService.is_email_in_freeze(user_info.email):
                 raise AccountRegisterError(
@@ -220,4 +225,4 @@ def _generate_account(provider: str, user_info: OAuthUserInfo):
     # Link account
     AccountService.link_account_integrate(provider, user_info.id, account)
 
-    return account
+    return account, oauth_new_user

api/controllers/console/billing/billing.py

@@ -1,8 +1,9 @@
 import base64
+from typing import Literal
 
 from flask import request
 from flask_restx import Resource, fields
-from pydantic import BaseModel, Field, field_validator
+from pydantic import BaseModel, Field
 from werkzeug.exceptions import BadRequest
 
 from controllers.console import console_ns
@@ -15,22 +16,8 @@ DEFAULT_REF_TEMPLATE_SWAGGER_2_0 = "#/definitions/{model}"
 
 
 class SubscriptionQuery(BaseModel):
-    plan: str = Field(..., description="Subscription plan")
-    interval: str = Field(..., description="Billing interval")
-
-    @field_validator("plan")
-    @classmethod
-    def validate_plan(cls, value: str) -> str:
-        if value not in [CloudPlan.PROFESSIONAL, CloudPlan.TEAM]:
-            raise ValueError("Invalid plan")
-        return value
-
-    @field_validator("interval")
-    @classmethod
-    def validate_interval(cls, value: str) -> str:
-        if value not in {"month", "year"}:
-            raise ValueError("Invalid interval")
-        return value
+    plan: Literal[CloudPlan.PROFESSIONAL, CloudPlan.TEAM] = Field(..., description="Subscription plan")
+    interval: Literal["month", "year"] = Field(..., description="Billing interval")
 
 
 class PartnerTenantsPayload(BaseModel):

api/controllers/console/datasets/datasets_segments.py

@@ -3,10 +3,12 @@ import uuid
 from flask import request
 from flask_restx import Resource, marshal
 from pydantic import BaseModel, Field
-from sqlalchemy import select
+from sqlalchemy import String, cast, func, or_, select
+from sqlalchemy.dialects.postgresql import JSONB
 from werkzeug.exceptions import Forbidden, NotFound
 
 import services
+from configs import dify_config
 from controllers.common.schema import register_schema_models
 from controllers.console import console_ns
 from controllers.console.app.error import ProviderNotInitializeError
@@ -143,7 +145,29 @@ class DatasetDocumentSegmentListApi(Resource):
             query = query.where(DocumentSegment.hit_count >= hit_count_gte)
 
         if keyword:
-            query = query.where(DocumentSegment.content.ilike(f"%{keyword}%"))
+            # Search in both content and keywords fields
+            # Use database-specific methods for JSON array search
+            if dify_config.SQLALCHEMY_DATABASE_URI_SCHEME == "postgresql":
+                # PostgreSQL: Use jsonb_array_elements_text to properly handle Unicode/Chinese text
+                keywords_condition = func.array_to_string(
+                    func.array(
+                        select(func.jsonb_array_elements_text(cast(DocumentSegment.keywords, JSONB)))
+                        .correlate(DocumentSegment)
+                        .scalar_subquery()
+                    ),
+                    ",",
+                ).ilike(f"%{keyword}%")
+            else:
+                # MySQL: Cast JSON to string for pattern matching
+                # MySQL stores Chinese text directly in JSON without Unicode escaping
+                keywords_condition = cast(DocumentSegment.keywords, String).ilike(f"%{keyword}%")
+
+            query = query.where(
+                or_(
+                    DocumentSegment.content.ilike(f"%{keyword}%"),
+                    keywords_condition,
+                )
+            )
 
         if args.enabled.lower() != "all":
             if args.enabled.lower() == "true":

api/controllers/console/explore/message.py

@@ -1,6 +1,5 @@
 import logging
 from typing import Literal
-from uuid import UUID
 
 from flask import request
 from flask_restx import marshal_with
@@ -26,6 +25,7 @@ from core.errors.error import ModelCurrentlyNotSupportError, ProviderTokenNotIni
 from core.model_runtime.errors.invoke import InvokeError
 from fields.message_fields import message_infinite_scroll_pagination_fields
 from libs import helper
+from libs.helper import UUIDStrOrEmpty
 from libs.login import current_account_with_tenant
 from models.model import AppMode
 from services.app_generate_service import AppGenerateService
@@ -44,8 +44,8 @@ logger = logging.getLogger(__name__)
 
 
 class MessageListQuery(BaseModel):
-    conversation_id: UUID
-    first_id: UUID | None = None
+    conversation_id: UUIDStrOrEmpty
+    first_id: UUIDStrOrEmpty | None = None
     limit: int = Field(default=20, ge=1, le=100)
 
 

api/controllers/console/explore/parameter.py

@@ -1,5 +1,3 @@
-from flask_restx import marshal_with
-
 from controllers.common import fields
 from controllers.console import console_ns
 from controllers.console.app.error import AppUnavailableError
@@ -13,7 +11,6 @@ from services.app_service import AppService
 class AppParameterApi(InstalledAppResource):
     """Resource for app variables."""
 
-    @marshal_with(fields.parameters_fields)
     def get(self, installed_app: InstalledApp):
         """Retrieve app parameters."""
         app_model = installed_app.app
@@ -37,7 +34,8 @@ class AppParameterApi(InstalledAppResource):
 
             user_input_form = features_dict.get("user_input_form", [])
 
-        return get_parameters_from_feature_dict(features_dict=features_dict, user_input_form=user_input_form)
+        parameters = get_parameters_from_feature_dict(features_dict=features_dict, user_input_form=user_input_form)
+        return fields.Parameters.model_validate(parameters).model_dump(mode="json")
 
 
 @console_ns.route("/installed-apps/<uuid:installed_app_id>/meta", endpoint="installed_app_meta")

api/controllers/console/explore/saved_message.py

@@ -1,5 +1,3 @@
-from uuid import UUID
-
 from flask import request
 from flask_restx import fields, marshal_with
 from pydantic import BaseModel, Field
@@ -10,19 +8,19 @@ from controllers.console import console_ns
 from controllers.console.explore.error import NotCompletionAppError
 from controllers.console.explore.wraps import InstalledAppResource
 from fields.conversation_fields import message_file_fields
-from libs.helper import TimestampField
+from libs.helper import TimestampField, UUIDStrOrEmpty
 from libs.login import current_account_with_tenant
 from services.errors.message import MessageNotExistsError
 from services.saved_message_service import SavedMessageService
 
 
 class SavedMessageListQuery(BaseModel):
-    last_id: UUID | None = None
+    last_id: UUIDStrOrEmpty | None = None
     limit: int = Field(default=20, ge=1, le=100)
 
 
 class SavedMessageCreatePayload(BaseModel):
-    message_id: UUID
+    message_id: UUIDStrOrEmpty
 
 
 register_schema_models(console_ns, SavedMessageListQuery, SavedMessageCreatePayload)

api/controllers/console/workspace/load_balancing_config.py

@@ -1,6 +1,8 @@
-from flask_restx import Resource, reqparse
+from flask_restx import Resource
+from pydantic import BaseModel
 from werkzeug.exceptions import Forbidden
 
+from controllers.common.schema import register_schema_models
 from controllers.console import console_ns
 from controllers.console.wraps import account_initialization_required, setup_required
 from core.model_runtime.entities.model_entities import ModelType
@@ -10,10 +12,20 @@ from models import TenantAccountRole
 from services.model_load_balancing_service import ModelLoadBalancingService
 
 
+class LoadBalancingCredentialPayload(BaseModel):
+    model: str
+    model_type: ModelType
+    credentials: dict[str, object]
+
+
+register_schema_models(console_ns, LoadBalancingCredentialPayload)
+
+
 @console_ns.route(
     "/workspaces/current/model-providers/<path:provider>/models/load-balancing-configs/credentials-validate"
 )
 class LoadBalancingCredentialsValidateApi(Resource):
+    @console_ns.expect(console_ns.models[LoadBalancingCredentialPayload.__name__])
     @setup_required
     @login_required
     @account_initialization_required
@@ -24,20 +36,7 @@ class LoadBalancingCredentialsValidateApi(Resource):
 
         tenant_id = current_tenant_id
 
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("model", type=str, required=True, nullable=False, location="json")
-            .add_argument(
-                "model_type",
-                type=str,
-                required=True,
-                nullable=False,
-                choices=[mt.value for mt in ModelType],
-                location="json",
-            )
-            .add_argument("credentials", type=dict, required=True, nullable=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = LoadBalancingCredentialPayload.model_validate(console_ns.payload or {})
 
         # validate model load balancing credentials
         model_load_balancing_service = ModelLoadBalancingService()
@@ -49,9 +48,9 @@ class LoadBalancingCredentialsValidateApi(Resource):
             model_load_balancing_service.validate_load_balancing_credentials(
                 tenant_id=tenant_id,
                 provider=provider,
-                model=args["model"],
-                model_type=args["model_type"],
-                credentials=args["credentials"],
+                model=payload.model,
+                model_type=payload.model_type,
+                credentials=payload.credentials,
             )
         except CredentialsValidateFailedError as ex:
             result = False
@@ -69,6 +68,7 @@ class LoadBalancingCredentialsValidateApi(Resource):
     "/workspaces/current/model-providers/<path:provider>/models/load-balancing-configs/<string:config_id>/credentials-validate"
 )
 class LoadBalancingConfigCredentialsValidateApi(Resource):
+    @console_ns.expect(console_ns.models[LoadBalancingCredentialPayload.__name__])
     @setup_required
     @login_required
     @account_initialization_required
@@ -79,20 +79,7 @@ class LoadBalancingConfigCredentialsValidateApi(Resource):
 
         tenant_id = current_tenant_id
 
-        parser = (
-            reqparse.RequestParser()
-            .add_argument("model", type=str, required=True, nullable=False, location="json")
-            .add_argument(
-                "model_type",
-                type=str,
-                required=True,
-                nullable=False,
-                choices=[mt.value for mt in ModelType],
-                location="json",
-            )
-            .add_argument("credentials", type=dict, required=True, nullable=False, location="json")
-        )
-        args = parser.parse_args()
+        payload = LoadBalancingCredentialPayload.model_validate(console_ns.payload or {})
 
         # validate model load balancing config credentials
         model_load_balancing_service = ModelLoadBalancingService()
@@ -104,9 +91,9 @@ class LoadBalancingConfigCredentialsValidateApi(Resource):
             model_load_balancing_service.validate_load_balancing_credentials(
                 tenant_id=tenant_id,
                 provider=provider,
-                model=args["model"],
-                model_type=args["model_type"],
-                credentials=args["credentials"],
+                model=payload.model,
+                model_type=payload.model_type,
+                credentials=payload.credentials,
                 config_id=config_id,
             )
         except CredentialsValidateFailedError as ex:

api/controllers/console/workspace/tool_providers.py

@@ -1,4 +1,5 @@
 import io
+import logging
 from urllib.parse import urlparse
 
 from flask import make_response, redirect, request, send_file
@@ -17,8 +18,8 @@ from controllers.console.wraps import (
     is_admin_or_owner_required,
     setup_required,
 )
+from core.db.session_factory import session_factory
 from core.entities.mcp_provider import MCPAuthentication, MCPConfiguration
-from core.helper.tool_provider_cache import ToolProviderListCache
 from core.mcp.auth.auth_flow import auth, handle_callback
 from core.mcp.error import MCPAuthError, MCPError, MCPRefreshTokenError
 from core.mcp.mcp_client import MCPClient
@@ -40,6 +41,8 @@ from services.tools.tools_manage_service import ToolCommonService
 from services.tools.tools_transform_service import ToolTransformService
 from services.tools.workflow_tools_manage_service import WorkflowToolManageService
 
+logger = logging.getLogger(__name__)
+
 
 def is_valid_url(url: str) -> bool:
     if not url:
@@ -945,8 +948,8 @@ class ToolProviderMCPApi(Resource):
         configuration = MCPConfiguration.model_validate(args["configuration"])
         authentication = MCPAuthentication.model_validate(args["authentication"]) if args["authentication"] else None
 
-        # Create provider in transaction
-        with Session(db.engine) as session, session.begin():
+        # 1) Create provider in a short transaction (no network I/O inside)
+        with session_factory.create_session() as session, session.begin():
             service = MCPToolManageService(session=session)
             result = service.create_provider(
                 tenant_id=tenant_id,
@@ -962,8 +965,26 @@ class ToolProviderMCPApi(Resource):
                 authentication=authentication,
             )
 
-        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
-        ToolProviderListCache.invalidate_cache(tenant_id)
+        # 2) Try to fetch tools immediately after creation so they appear without a second save.
+        #    Perform network I/O outside any DB session to avoid holding locks.
+        try:
+            reconnect = MCPToolManageService.reconnect_with_url(
+                server_url=args["server_url"],
+                headers=args.get("headers") or {},
+                timeout=configuration.timeout,
+                sse_read_timeout=configuration.sse_read_timeout,
+            )
+            # Update just-created provider with authed/tools in a new short transaction
+            with session_factory.create_session() as session, session.begin():
+                service = MCPToolManageService(session=session)
+                db_provider = service.get_provider(provider_id=result.id, tenant_id=tenant_id)
+                db_provider.authed = reconnect.authed
+                db_provider.tools = reconnect.tools
+
+                result = ToolTransformService.mcp_provider_to_user_provider(db_provider, for_list=True)
+        except Exception:
+            # Best-effort: if initial fetch fails (e.g., auth required), return created provider as-is
+            logger.warning("Failed to fetch MCP tools after creation", exc_info=True)
 
         return jsonable_encoder(result)
 
@@ -1011,9 +1032,6 @@ class ToolProviderMCPApi(Resource):
                 validation_result=validation_result,
             )
 
-        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
-        ToolProviderListCache.invalidate_cache(current_tenant_id)
-
         return {"result": "success"}
 
     @console_ns.expect(parser_mcp_delete)
@@ -1028,9 +1046,6 @@ class ToolProviderMCPApi(Resource):
             service = MCPToolManageService(session=session)
             service.delete_provider(tenant_id=current_tenant_id, provider_id=args["provider_id"])
 
-        # Invalidate cache AFTER transaction commits to avoid holding locks during Redis operations
-        ToolProviderListCache.invalidate_cache(current_tenant_id)
-
         return {"result": "success"}
 
 
@@ -1081,8 +1096,6 @@ class ToolMCPAuthApi(Resource):
                         credentials=provider_entity.credentials,
                         authed=True,
                     )
-                # Invalidate cache after updating credentials
-                ToolProviderListCache.invalidate_cache(tenant_id)
                 return {"result": "success"}
         except MCPAuthError as e:
             try:
@@ -1096,22 +1109,16 @@ class ToolMCPAuthApi(Resource):
                 with Session(db.engine) as session, session.begin():
                     service = MCPToolManageService(session=session)
                     response = service.execute_auth_actions(auth_result)
-                    # Invalidate cache after auth actions may have updated provider state
-                    ToolProviderListCache.invalidate_cache(tenant_id)
                     return response
             except MCPRefreshTokenError as e:
                 with Session(db.engine) as session, session.begin():
                     service = MCPToolManageService(session=session)
                     service.clear_provider_credentials(provider_id=provider_id, tenant_id=tenant_id)
-                # Invalidate cache after clearing credentials
-                ToolProviderListCache.invalidate_cache(tenant_id)
                 raise ValueError(f"Failed to refresh token, please try to authorize again: {e}") from e
         except (MCPError, ValueError) as e:
             with Session(db.engine) as session, session.begin():
                 service = MCPToolManageService(session=session)
                 service.clear_provider_credentials(provider_id=provider_id, tenant_id=tenant_id)
-            # Invalidate cache after clearing credentials
-            ToolProviderListCache.invalidate_cache(tenant_id)
             raise ValueError(f"Failed to connect to MCP server: {e}") from e
 
 

api/controllers/service_api/app/annotation.py

@@ -1,7 +1,7 @@
 from typing import Literal
 
 from flask import request
-from flask_restx import Api, Namespace, Resource, fields
+from flask_restx import Namespace, Resource, fields
 from flask_restx.api import HTTPStatus
 from pydantic import BaseModel, Field
 
@@ -92,7 +92,7 @@ annotation_list_fields = {
 }
 
 
-def build_annotation_list_model(api_or_ns: Api | Namespace):
+def build_annotation_list_model(api_or_ns: Namespace):
     """Build the annotation list model for the API or Namespace."""
     copied_annotation_list_fields = annotation_list_fields.copy()
     copied_annotation_list_fields["data"] = fields.List(fields.Nested(build_annotation_model(api_or_ns)))

api/controllers/service_api/app/app.py

@@ -1,6 +1,6 @@
 from flask_restx import Resource
 
-from controllers.common.fields import build_parameters_model
+from controllers.common.fields import Parameters
 from controllers.service_api import service_api_ns
 from controllers.service_api.app.error import AppUnavailableError
 from controllers.service_api.wraps import validate_app_token
@@ -23,7 +23,6 @@ class AppParameterApi(Resource):
         }
     )
     @validate_app_token
-    @service_api_ns.marshal_with(build_parameters_model(service_api_ns))
     def get(self, app_model: App):
         """Retrieve app parameters.
 
@@ -45,7 +44,8 @@ class AppParameterApi(Resource):
 
             user_input_form = features_dict.get("user_input_form", [])
 
-        return get_parameters_from_feature_dict(features_dict=features_dict, user_input_form=user_input_form)
+        parameters = get_parameters_from_feature_dict(features_dict=features_dict, user_input_form=user_input_form)
+        return Parameters.model_validate(parameters).model_dump(mode="json")
 
 
 @service_api_ns.route("/meta")

api/controllers/service_api/app/site.py

@@ -1,7 +1,7 @@
 from flask_restx import Resource
 from werkzeug.exceptions import Forbidden
 
-from controllers.common.fields import build_site_model
+from controllers.common.fields import Site as SiteResponse
 from controllers.service_api import service_api_ns
 from controllers.service_api.wraps import validate_app_token
 from extensions.ext_database import db
@@ -23,7 +23,6 @@ class AppSiteApi(Resource):
         }
     )
     @validate_app_token
-    @service_api_ns.marshal_with(build_site_model(service_api_ns))
     def get(self, app_model: App):
         """Retrieve app site info.
 
@@ -38,4 +37,4 @@ class AppSiteApi(Resource):
         if app_model.tenant.status == TenantStatus.ARCHIVE:
             raise Forbidden()
 
-        return site
+        return SiteResponse.model_validate(site).model_dump(mode="json")

api/controllers/service_api/app/workflow.py

@@ -3,7 +3,7 @@ from typing import Any, Literal
 
 from dateutil.parser import isoparse
 from flask import request
-from flask_restx import Api, Namespace, Resource, fields
+from flask_restx import Namespace, Resource, fields
 from pydantic import BaseModel, Field
 from sqlalchemy.orm import Session, sessionmaker
 from werkzeug.exceptions import BadRequest, InternalServerError, NotFound
@@ -78,7 +78,7 @@ workflow_run_fields = {
 }
 
 
-def build_workflow_run_model(api_or_ns: Api | Namespace):
+def build_workflow_run_model(api_or_ns: Namespace):
     """Build the workflow run model for the API or Namespace."""
     return api_or_ns.model("WorkflowRun", workflow_run_fields)
 

api/controllers/service_api/dataset/dataset.py

@@ -13,7 +13,6 @@ from controllers.service_api.dataset.error import DatasetInUseError, DatasetName
 from controllers.service_api.wraps import (
     DatasetApiResource,
     cloud_edition_billing_rate_limit_check,
-    validate_dataset_token,
 )
 from core.model_runtime.entities.model_entities import ModelType
 from core.provider_manager import ProviderManager
@@ -460,9 +459,8 @@ class DatasetTagsApi(DatasetApiResource):
             401: "Unauthorized - invalid API token",
         }
     )
-    @validate_dataset_token
     @service_api_ns.marshal_with(build_dataset_tag_fields(service_api_ns))
-    def get(self, _, dataset_id):
+    def get(self, _):
         """Get all knowledge type tags."""
         assert isinstance(current_user, Account)
         cid = current_user.current_tenant_id
@@ -482,8 +480,7 @@ class DatasetTagsApi(DatasetApiResource):
         }
     )
     @service_api_ns.marshal_with(build_dataset_tag_fields(service_api_ns))
-    @validate_dataset_token
-    def post(self, _, dataset_id):
+    def post(self, _):
         """Add a knowledge type tag."""
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
@@ -506,8 +503,7 @@ class DatasetTagsApi(DatasetApiResource):
         }
     )
     @service_api_ns.marshal_with(build_dataset_tag_fields(service_api_ns))
-    @validate_dataset_token
-    def patch(self, _, dataset_id):
+    def patch(self, _):
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
             raise Forbidden()
@@ -533,9 +529,8 @@ class DatasetTagsApi(DatasetApiResource):
             403: "Forbidden - insufficient permissions",
         }
     )
-    @validate_dataset_token
     @edit_permission_required
-    def delete(self, _, dataset_id):
+    def delete(self, _):
         """Delete a knowledge type tag."""
         payload = TagDeletePayload.model_validate(service_api_ns.payload or {})
         TagService.delete_tag(payload.tag_id)
@@ -555,8 +550,7 @@ class DatasetTagBindingApi(DatasetApiResource):
             403: "Forbidden - insufficient permissions",
         }
     )
-    @validate_dataset_token
-    def post(self, _, dataset_id):
+    def post(self, _):
         # The role of the current user in the ta table must be admin, owner, editor, or dataset_operator
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
@@ -580,8 +574,7 @@ class DatasetTagUnbindingApi(DatasetApiResource):
             403: "Forbidden - insufficient permissions",
         }
     )
-    @validate_dataset_token
-    def post(self, _, dataset_id):
+    def post(self, _):
         # The role of the current user in the ta table must be admin, owner, editor, or dataset_operator
         assert isinstance(current_user, Account)
         if not (current_user.has_edit_permission or current_user.is_dataset_editor):
@@ -604,7 +597,6 @@ class DatasetTagsBindingStatusApi(DatasetApiResource):
             401: "Unauthorized - invalid API token",
         }
     )
-    @validate_dataset_token
     def get(self, _, *args, **kwargs):
         """Get all knowledge type tags."""
         dataset_id = kwargs.get("dataset_id")

api/controllers/web/app.py

@@ -1,7 +1,7 @@
 import logging
 
 from flask import request
-from flask_restx import Resource, marshal_with
+from flask_restx import Resource
 from pydantic import BaseModel, ConfigDict, Field
 from werkzeug.exceptions import Unauthorized
 
@@ -50,7 +50,6 @@ class AppParameterApi(WebApiResource):
             500: "Internal Server Error",
         }
     )
-    @marshal_with(fields.parameters_fields)
     def get(self, app_model: App, end_user):
         """Retrieve app parameters."""
         if app_model.mode in {AppMode.ADVANCED_CHAT, AppMode.WORKFLOW}:
@@ -69,7 +68,8 @@ class AppParameterApi(WebApiResource):
 
             user_input_form = features_dict.get("user_input_form", [])
 
-        return get_parameters_from_feature_dict(features_dict=features_dict, user_input_form=user_input_form)
+        parameters = get_parameters_from_feature_dict(features_dict=features_dict, user_input_form=user_input_form)
+        return fields.Parameters.model_validate(parameters).model_dump(mode="json")
 
 
 @web_ns.route("/meta")

api/core/agent/cot_agent_runner.py

@@ -22,6 +22,7 @@ from core.prompt.agent_history_prompt_transform import AgentHistoryPromptTransfo
 from core.tools.__base.tool import Tool
 from core.tools.entities.tool_entities import ToolInvokeMeta
 from core.tools.tool_engine import ToolEngine
+from core.workflow.nodes.agent.exc import AgentMaxIterationError
 from models.model import Message
 
 logger = logging.getLogger(__name__)
@@ -165,6 +166,11 @@ class CotAgentRunner(BaseAgentRunner, ABC):
             scratchpad.thought = scratchpad.thought.strip() or "I am thinking about how to help you"
             self._agent_scratchpad.append(scratchpad)
 
+            # Check if max iteration is reached and model still wants to call tools
+            if iteration_step == max_iteration_steps and scratchpad.action:
+                if scratchpad.action.action_name.lower() != "final answer":
+                    raise AgentMaxIterationError(app_config.agent.max_iteration)
+
             # get llm usage
             if "usage" in usage_dict:
                 if usage_dict["usage"] is not None:

api/core/agent/fc_agent_runner.py

@@ -25,6 +25,7 @@ from core.model_runtime.entities.message_entities import ImagePromptMessageConte
 from core.prompt.agent_history_prompt_transform import AgentHistoryPromptTransform
 from core.tools.entities.tool_entities import ToolInvokeMeta
 from core.tools.tool_engine import ToolEngine
+from core.workflow.nodes.agent.exc import AgentMaxIterationError
 from models.model import Message
 
 logger = logging.getLogger(__name__)
@@ -222,6 +223,10 @@ class FunctionCallAgentRunner(BaseAgentRunner):
 
             final_answer += response + "\n"
 
+            # Check if max iteration is reached and model still wants to call tools
+            if iteration_step == max_iteration_steps and tool_calls:
+                raise AgentMaxIterationError(app_config.agent.max_iteration)
+
             # call tools
             tool_responses = []
             for tool_call_id, tool_call_name, tool_call_args in tool_calls:

api/core/app/apps/base_app_queue_manager.py

@@ -90,6 +90,7 @@ class AppQueueManager:
         """
         self._clear_task_belong_cache()
         self._q.put(None)
+        self._graph_runtime_state = None  # Release reference to allow GC to reclaim memory
 
     def _clear_task_belong_cache(self) -> None:
         """

api/core/entities/model_entities.py

@@ -30,7 +30,6 @@ class SimpleModelProviderEntity(BaseModel):
     label: I18nObject
     icon_small: I18nObject | None = None
     icon_small_dark: I18nObject | None = None
-    icon_large: I18nObject | None = None
     supported_model_types: list[ModelType]
 
     def __init__(self, provider_entity: ProviderEntity):
@@ -44,7 +43,6 @@ class SimpleModelProviderEntity(BaseModel):
             label=provider_entity.label,
             icon_small=provider_entity.icon_small,
             icon_small_dark=provider_entity.icon_small_dark,
-            icon_large=provider_entity.icon_large,
             supported_model_types=provider_entity.supported_model_types,
         )
 
@@ -94,7 +92,6 @@ class DefaultModelProviderEntity(BaseModel):
     provider: str
     label: I18nObject
     icon_small: I18nObject | None = None
-    icon_large: I18nObject | None = None
     supported_model_types: Sequence[ModelType] = []
 
 

api/core/helper/code_executor/jinja2/jinja2_transformer.py

@@ -1,9 +1,14 @@
+from collections.abc import Mapping
 from textwrap import dedent
+from typing import Any
 
 from core.helper.code_executor.template_transformer import TemplateTransformer
 
 
 class Jinja2TemplateTransformer(TemplateTransformer):
+    # Use separate placeholder for base64-encoded template to avoid confusion
+    _template_b64_placeholder: str = "{{template_b64}}"
+
     @classmethod
     def transform_response(cls, response: str):
         """
@@ -13,18 +18,35 @@ class Jinja2TemplateTransformer(TemplateTransformer):
         """
         return {"result": cls.extract_result_str_from_response(response)}
 
+    @classmethod
+    def assemble_runner_script(cls, code: str, inputs: Mapping[str, Any]) -> str:
+        """
+        Override base class to use base64 encoding for template code.
+        This prevents issues with special characters (quotes, newlines) in templates
+        breaking the generated Python script. Fixes #26818.
+        """
+        script = cls.get_runner_script()
+        # Encode template as base64 to safely embed any content including quotes
+        code_b64 = cls.serialize_code(code)
+        script = script.replace(cls._template_b64_placeholder, code_b64)
+        inputs_str = cls.serialize_inputs(inputs)
+        script = script.replace(cls._inputs_placeholder, inputs_str)
+        return script
+
     @classmethod
     def get_runner_script(cls) -> str:
         runner_script = dedent(f"""
-            # declare main function
-            def main(**inputs):
-                import jinja2
-                template = jinja2.Template('''{cls._code_placeholder}''')
-                return template.render(**inputs)
-
+            import jinja2
             import json
             from base64 import b64decode
 
+            # declare main function
+            def main(**inputs):
+                # Decode base64-encoded template to handle special characters safely
+                template_code = b64decode('{cls._template_b64_placeholder}').decode('utf-8')
+                template = jinja2.Template(template_code)
+                return template.render(**inputs)
+
             # decode and prepare input dict
             inputs_obj = json.loads(b64decode('{cls._inputs_placeholder}').decode('utf-8'))
 

api/core/helper/code_executor/template_transformer.py

@@ -13,6 +13,15 @@ class TemplateTransformer(ABC):
     _inputs_placeholder: str = "{{inputs}}"
     _result_tag: str = "<<RESULT>>"
 
+    @classmethod
+    def serialize_code(cls, code: str) -> str:
+        """
+        Serialize template code to base64 to safely embed in generated script.
+        This prevents issues with special characters like quotes breaking the script.
+        """
+        code_bytes = code.encode("utf-8")
+        return b64encode(code_bytes).decode("utf-8")
+
     @classmethod
     def transform_caller(cls, code: str, inputs: Mapping[str, Any]) -> tuple[str, str]:
         """

api/core/helper/ssrf_proxy.py

@@ -88,7 +88,41 @@ def _get_user_provided_host_header(headers: dict | None) -> str | None:
     return None
 
 
+def _inject_trace_headers(headers: dict | None) -> dict:
+    """
+    Inject W3C traceparent header for distributed tracing.
+
+    When OTEL is enabled, HTTPXClientInstrumentor handles trace propagation automatically.
+    When OTEL is disabled, we manually inject the traceparent header.
+    """
+    if headers is None:
+        headers = {}
+
+    # Skip if already present (case-insensitive check)
+    for key in headers:
+        if key.lower() == "traceparent":
+            return headers
+
+    # Skip if OTEL is enabled - HTTPXClientInstrumentor handles this automatically
+    if dify_config.ENABLE_OTEL:
+        return headers
+
+    # Generate and inject traceparent for non-OTEL scenarios
+    try:
+        from core.helper.trace_id_helper import generate_traceparent_header
+
+        traceparent = generate_traceparent_header()
+        if traceparent:
+            headers["traceparent"] = traceparent
+    except Exception:
+        # Silently ignore errors to avoid breaking requests
+        logger.debug("Failed to generate traceparent header", exc_info=True)
+
+    return headers
+
+
 def make_request(method, url, max_retries=SSRF_DEFAULT_MAX_RETRIES, **kwargs):
+    # Convert requests-style allow_redirects to httpx-style follow_redirects
     if "allow_redirects" in kwargs:
         allow_redirects = kwargs.pop("allow_redirects")
         if "follow_redirects" not in kwargs:
@@ -106,18 +140,21 @@ def make_request(method, url, max_retries=SSRF_DEFAULT_MAX_RETRIES, **kwargs):
     verify_option = kwargs.pop("ssl_verify", dify_config.HTTP_REQUEST_NODE_SSL_VERIFY)
     client = _get_ssrf_client(verify_option)
 
+    # Inject traceparent header for distributed tracing (when OTEL is not enabled)
+    headers = kwargs.get("headers") or {}
+    headers = _inject_trace_headers(headers)
+    kwargs["headers"] = headers
+
     # Preserve user-provided Host header
     # When using a forward proxy, httpx may override the Host header based on the URL.
     # We extract and preserve any explicitly set Host header to support virtual hosting.
-    headers = kwargs.get("headers", {})
     user_provided_host = _get_user_provided_host_header(headers)
 
     retries = 0
     while retries <= max_retries:
         try:
-            # Build the request manually to preserve the Host header
-            # httpx may override the Host header when using a proxy, so we use
-            # the request API to explicitly set headers before sending
+            # Preserve the user-provided Host header
+            # httpx may override the Host header when using a proxy
             headers = {k: v for k, v in headers.items() if k.lower() != "host"}
             if user_provided_host is not None:
                 headers["host"] = user_provided_host

api/core/helper/tool_provider_cache.py

@@ -1,56 +0,0 @@
-import json
-import logging
-from typing import Any
-
-from core.tools.entities.api_entities import ToolProviderTypeApiLiteral
-from extensions.ext_redis import redis_client, redis_fallback
-
-logger = logging.getLogger(__name__)
-
-
-class ToolProviderListCache:
-    """Cache for tool provider lists"""
-
-    CACHE_TTL = 300  # 5 minutes
-
-    @staticmethod
-    def _generate_cache_key(tenant_id: str, typ: ToolProviderTypeApiLiteral = None) -> str:
-        """Generate cache key for tool providers list"""
-        type_filter = typ or "all"
-        return f"tool_providers:tenant_id:{tenant_id}:type:{type_filter}"
-
-    @staticmethod
-    @redis_fallback(default_return=None)
-    def get_cached_providers(tenant_id: str, typ: ToolProviderTypeApiLiteral = None) -> list[dict[str, Any]] | None:
-        """Get cached tool providers"""
-        cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-        cached_data = redis_client.get(cache_key)
-        if cached_data:
-            try:
-                return json.loads(cached_data.decode("utf-8"))
-            except (json.JSONDecodeError, UnicodeDecodeError):
-                logger.warning("Failed to decode cached tool providers data")
-                return None
-        return None
-
-    @staticmethod
-    @redis_fallback()
-    def set_cached_providers(tenant_id: str, typ: ToolProviderTypeApiLiteral, providers: list[dict[str, Any]]):
-        """Cache tool providers"""
-        cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-        redis_client.setex(cache_key, ToolProviderListCache.CACHE_TTL, json.dumps(providers))
-
-    @staticmethod
-    @redis_fallback()
-    def invalidate_cache(tenant_id: str, typ: ToolProviderTypeApiLiteral = None):
-        """Invalidate cache for tool providers"""
-        if typ:
-            # Invalidate specific type cache
-            cache_key = ToolProviderListCache._generate_cache_key(tenant_id, typ)
-            redis_client.delete(cache_key)
-        else:
-            # Invalidate all caches for this tenant
-            pattern = f"tool_providers:tenant_id:{tenant_id}:*"
-            keys = list(redis_client.scan_iter(pattern))
-            if keys:
-                redis_client.delete(*keys)

api/core/helper/trace_id_helper.py

@@ -103,3 +103,60 @@ def parse_traceparent_header(traceparent: str) -> str | None:
         if len(parts) == 4 and len(parts[1]) == 32:
             return parts[1]
     return None
+
+
+def get_span_id_from_otel_context() -> str | None:
+    """
+    Retrieve the current span ID from the active OpenTelemetry trace context.
+
+    Returns:
+        A 16-character hex string representing the span ID, or None if not available.
+    """
+    try:
+        from opentelemetry.trace import get_current_span
+        from opentelemetry.trace.span import INVALID_SPAN_ID
+
+        span = get_current_span()
+        if not span:
+            return None
+
+        span_context = span.get_span_context()
+        if not span_context or span_context.span_id == INVALID_SPAN_ID:
+            return None
+
+        return f"{span_context.span_id:016x}"
+    except Exception:
+        return None
+
+
+def generate_traceparent_header() -> str | None:
+    """
+    Generate a W3C traceparent header from the current context.
+
+    Uses OpenTelemetry context if available, otherwise uses the
+    ContextVar-based trace_id from the logging context.
+
+    Format: {version}-{trace_id}-{span_id}-{flags}
+    Example: 00-5b8aa5a2d2c872e8321cf37308d69df2-051581bf3bb55c45-01
+
+    Returns:
+        A valid traceparent header string, or None if generation fails.
+    """
+    import uuid
+
+    # Try OTEL context first
+    trace_id = get_trace_id_from_otel_context()
+    span_id = get_span_id_from_otel_context()
+
+    if trace_id and span_id:
+        return f"00-{trace_id}-{span_id}-01"
+
+    # Fallback: use ContextVar-based trace_id or generate new one
+    from core.logging.context import get_trace_id as get_logging_trace_id
+
+    trace_id = get_logging_trace_id() or uuid.uuid4().hex
+
+    # Generate a new span_id (16 hex chars)
+    span_id = uuid.uuid4().hex[:16]
+
+    return f"00-{trace_id}-{span_id}-01"

api/core/logging/__init__.py

@@ -0,0 +1,20 @@
+"""Structured logging components for Dify."""
+
+from core.logging.context import (
+    clear_request_context,
+    get_request_id,
+    get_trace_id,
+    init_request_context,
+)
+from core.logging.filters import IdentityContextFilter, TraceContextFilter
+from core.logging.structured_formatter import StructuredJSONFormatter
+
+__all__ = [
+    "IdentityContextFilter",
+    "StructuredJSONFormatter",
+    "TraceContextFilter",
+    "clear_request_context",
+    "get_request_id",
+    "get_trace_id",
+    "init_request_context",
+]

api/core/logging/context.py

@@ -0,0 +1,35 @@
+"""Request context for logging - framework agnostic.
+
+This module provides request-scoped context variables for logging,
+using Python's contextvars for thread-safe and async-safe storage.
+"""
+
+import uuid
+from contextvars import ContextVar
+
+_request_id: ContextVar[str] = ContextVar("log_request_id", default="")
+_trace_id: ContextVar[str] = ContextVar("log_trace_id", default="")
+
+
+def get_request_id() -> str:
+    """Get current request ID (10 hex chars)."""
+    return _request_id.get()
+
+
+def get_trace_id() -> str:
+    """Get fallback trace ID when OTEL is unavailable (32 hex chars)."""
+    return _trace_id.get()
+
+
+def init_request_context() -> None:
+    """Initialize request context. Call at start of each request."""
+    req_id = uuid.uuid4().hex[:10]
+    trace_id = uuid.uuid5(uuid.NAMESPACE_DNS, req_id).hex
+    _request_id.set(req_id)
+    _trace_id.set(trace_id)
+
+
+def clear_request_context() -> None:
+    """Clear request context. Call at end of request (optional)."""
+    _request_id.set("")
+    _trace_id.set("")

api/core/logging/filters.py

@@ -0,0 +1,94 @@
+"""Logging filters for structured logging."""
+
+import contextlib
+import logging
+
+import flask
+
+from core.logging.context import get_request_id, get_trace_id
+
+
+class TraceContextFilter(logging.Filter):
+    """
+    Filter that adds trace_id and span_id to log records.
+    Integrates with OpenTelemetry when available, falls back to ContextVar-based trace_id.
+    """
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        # Get trace context from OpenTelemetry
+        trace_id, span_id = self._get_otel_context()
+
+        # Set trace_id (fallback to ContextVar if no OTEL context)
+        if trace_id:
+            record.trace_id = trace_id
+        else:
+            record.trace_id = get_trace_id()
+
+        record.span_id = span_id or ""
+
+        # For backward compatibility, also set req_id
+        record.req_id = get_request_id()
+
+        return True
+
+    def _get_otel_context(self) -> tuple[str, str]:
+        """Extract trace_id and span_id from OpenTelemetry context."""
+        with contextlib.suppress(Exception):
+            from opentelemetry.trace import get_current_span
+            from opentelemetry.trace.span import INVALID_SPAN_ID, INVALID_TRACE_ID
+
+            span = get_current_span()
+            if span and span.get_span_context():
+                ctx = span.get_span_context()
+                if ctx.is_valid and ctx.trace_id != INVALID_TRACE_ID:
+                    trace_id = f"{ctx.trace_id:032x}"
+                    span_id = f"{ctx.span_id:016x}" if ctx.span_id != INVALID_SPAN_ID else ""
+                    return trace_id, span_id
+        return "", ""
+
+
+class IdentityContextFilter(logging.Filter):
+    """
+    Filter that adds user identity context to log records.
+    Extracts tenant_id, user_id, and user_type from Flask-Login current_user.
+    """
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        identity = self._extract_identity()
+        record.tenant_id = identity.get("tenant_id", "")
+        record.user_id = identity.get("user_id", "")
+        record.user_type = identity.get("user_type", "")
+        return True
+
+    def _extract_identity(self) -> dict[str, str]:
+        """Extract identity from current_user if in request context."""
+        try:
+            if not flask.has_request_context():
+                return {}
+            from flask_login import current_user
+
+            # Check if user is authenticated using the proxy
+            if not current_user.is_authenticated:
+                return {}
+
+            # Access the underlying user object
+            user = current_user
+
+            from models import Account
+            from models.model import EndUser
+
+            identity: dict[str, str] = {}
+
+            if isinstance(user, Account):
+                if user.current_tenant_id:
+                    identity["tenant_id"] = user.current_tenant_id
+                identity["user_id"] = user.id
+                identity["user_type"] = "account"
+            elif isinstance(user, EndUser):
+                identity["tenant_id"] = user.tenant_id
+                identity["user_id"] = user.id
+                identity["user_type"] = user.type or "end_user"
+
+            return identity
+        except Exception:
+            return {}

api/core/logging/structured_formatter.py

@@ -0,0 +1,107 @@
+"""Structured JSON log formatter for Dify."""
+
+import logging
+import traceback
+from datetime import UTC, datetime
+from typing import Any
+
+import orjson
+
+from configs import dify_config
+
+
+class StructuredJSONFormatter(logging.Formatter):
+    """
+    JSON log formatter following the specified schema:
+    {
+      "ts": "ISO 8601 UTC",
+      "severity": "INFO|ERROR|WARN|DEBUG",
+      "service": "service name",
+      "caller": "file:line",
+      "trace_id": "hex 32",
+      "span_id": "hex 16",
+      "identity": { "tenant_id", "user_id", "user_type" },
+      "message": "log message",
+      "attributes": { ... },
+      "stack_trace": "..."
+    }
+    """
+
+    SEVERITY_MAP: dict[int, str] = {
+        logging.DEBUG: "DEBUG",
+        logging.INFO: "INFO",
+        logging.WARNING: "WARN",
+        logging.ERROR: "ERROR",
+        logging.CRITICAL: "ERROR",
+    }
+
+    def __init__(self, service_name: str | None = None):
+        super().__init__()
+        self._service_name = service_name or dify_config.APPLICATION_NAME
+
+    def format(self, record: logging.LogRecord) -> str:
+        log_dict = self._build_log_dict(record)
+        try:
+            return orjson.dumps(log_dict).decode("utf-8")
+        except TypeError:
+            # Fallback: convert non-serializable objects to string
+            import json
+
+            return json.dumps(log_dict, default=str, ensure_ascii=False)
+
+    def _build_log_dict(self, record: logging.LogRecord) -> dict[str, Any]:
+        # Core fields
+        log_dict: dict[str, Any] = {
+            "ts": datetime.now(UTC).isoformat(timespec="milliseconds").replace("+00:00", "Z"),
+            "severity": self.SEVERITY_MAP.get(record.levelno, "INFO"),
+            "service": self._service_name,
+            "caller": f"{record.filename}:{record.lineno}",
+            "message": record.getMessage(),
+        }
+
+        # Trace context (from TraceContextFilter)
+        trace_id = getattr(record, "trace_id", "")
+        span_id = getattr(record, "span_id", "")
+
+        if trace_id:
+            log_dict["trace_id"] = trace_id
+        if span_id:
+            log_dict["span_id"] = span_id
+
+        # Identity context (from IdentityContextFilter)
+        identity = self._extract_identity(record)
+        if identity:
+            log_dict["identity"] = identity
+
+        # Dynamic attributes
+        attributes = getattr(record, "attributes", None)
+        if attributes:
+            log_dict["attributes"] = attributes
+
+        # Stack trace for errors with exceptions
+        if record.exc_info and record.levelno >= logging.ERROR:
+            log_dict["stack_trace"] = self._format_exception(record.exc_info)
+
+        return log_dict
+
+    def _extract_identity(self, record: logging.LogRecord) -> dict[str, str] | None:
+        tenant_id = getattr(record, "tenant_id", None)
+        user_id = getattr(record, "user_id", None)
+        user_type = getattr(record, "user_type", None)
+
+        if not any([tenant_id, user_id, user_type]):
+            return None
+
+        identity: dict[str, str] = {}
+        if tenant_id:
+            identity["tenant_id"] = tenant_id
+        if user_id:
+            identity["user_id"] = user_id
+        if user_type:
+            identity["user_type"] = user_type
+        return identity
+
+    def _format_exception(self, exc_info: tuple[Any, ...]) -> str:
+        if exc_info and exc_info[0] is not None:
+            return "".join(traceback.format_exception(*exc_info))
+        return ""

api/core/mcp/client/streamable_client.py

@@ -313,17 +313,20 @@ class StreamableHTTPTransport:
             if is_initialization:
                 self._maybe_extract_session_id_from_response(response)
 
-            content_type = cast(str, response.headers.get(CONTENT_TYPE, "").lower())
+            # Per https://modelcontextprotocol.io/specification/2025-06-18/basic#notifications:
+            # The server MUST NOT send a response to notifications.
+            if isinstance(message.root, JSONRPCRequest):
+                content_type = cast(str, response.headers.get(CONTENT_TYPE, "").lower())
 
-            if content_type.startswith(JSON):
-                self._handle_json_response(response, ctx.server_to_client_queue)
-            elif content_type.startswith(SSE):
-                self._handle_sse_response(response, ctx)
-            else:
-                self._handle_unexpected_content_type(
-                    content_type,
-                    ctx.server_to_client_queue,
-                )
+                if content_type.startswith(JSON):
+                    self._handle_json_response(response, ctx.server_to_client_queue)
+                elif content_type.startswith(SSE):
+                    self._handle_sse_response(response, ctx)
+                else:
+                    self._handle_unexpected_content_type(
+                        content_type,
+                        ctx.server_to_client_queue,
+                    )
 
     def _handle_json_response(
         self,

api/core/model_runtime/entities/provider_entities.py

@@ -100,7 +100,6 @@ class SimpleProviderEntity(BaseModel):
     label: I18nObject
     icon_small: I18nObject | None = None
     icon_small_dark: I18nObject | None = None
-    icon_large: I18nObject | None = None
     supported_model_types: Sequence[ModelType]
     models: list[AIModelEntity] = []
 
@@ -123,7 +122,6 @@ class ProviderEntity(BaseModel):
     label: I18nObject
     description: I18nObject | None = None
     icon_small: I18nObject | None = None
-    icon_large: I18nObject | None = None
     icon_small_dark: I18nObject | None = None
     background: str | None = None
     help: ProviderHelpEntity | None = None
@@ -157,7 +155,6 @@ class ProviderEntity(BaseModel):
             provider=self.provider,
             label=self.label,
             icon_small=self.icon_small,
-            icon_large=self.icon_large,
             supported_model_types=self.supported_model_types,
             models=self.models,
         )

api/core/model_runtime/model_providers/model_provider_factory.py

@@ -285,7 +285,7 @@ class ModelProviderFactory:
         """
         Get provider icon
         :param provider: provider name
-        :param icon_type: icon type (icon_small or icon_large)
+        :param icon_type: icon type (icon_small or icon_small_dark)
         :param lang: language (zh_Hans or en_US)
         :return: provider icon
         """
@@ -309,13 +309,7 @@ class ModelProviderFactory:
             else:
                 file_name = provider_schema.icon_small_dark.en_US
         else:
-            if not provider_schema.icon_large:
-                raise ValueError(f"Provider {provider} does not have large icon.")
-
-            if lang.lower() == "zh_hans":
-                file_name = provider_schema.icon_large.zh_Hans
-            else:
-                file_name = provider_schema.icon_large.en_US
+            raise ValueError(f"Unsupported icon type: {icon_type}.")
 
         if not file_name:
             raise ValueError(f"Provider {provider} does not have icon.")

api/core/plugin/entities/parameters.py

@@ -76,7 +76,7 @@ class PluginParameter(BaseModel):
     auto_generate: PluginParameterAutoGenerate | None = None
     template: PluginParameterTemplate | None = None
     required: bool = False
-    default: Union[float, int, str, bool] | None = None
+    default: Union[float, int, str, bool, list, dict] | None = None
     min: Union[float, int] | None = None
     max: Union[float, int] | None = None
     precision: int | None = None

api/core/plugin/impl/base.py

@@ -103,6 +103,9 @@ class BasePluginClient:
         prepared_headers["X-Api-Key"] = dify_config.PLUGIN_DAEMON_KEY
         prepared_headers.setdefault("Accept-Encoding", "gzip, deflate, br")
 
+        # Inject traceparent header for distributed tracing
+        self._inject_trace_headers(prepared_headers)
+
         prepared_data: bytes | dict[str, Any] | str | None = (
             data if isinstance(data, (bytes, str, dict)) or data is None else None
         )
@@ -114,6 +117,31 @@ class BasePluginClient:
 
         return str(url), prepared_headers, prepared_data, params, files
 
+    def _inject_trace_headers(self, headers: dict[str, str]) -> None:
+        """
+        Inject W3C traceparent header for distributed tracing.
+
+        This ensures trace context is propagated to plugin daemon even if
+        HTTPXClientInstrumentor doesn't cover module-level httpx functions.
+        """
+        if not dify_config.ENABLE_OTEL:
+            return
+
+        import contextlib
+
+        # Skip if already present (case-insensitive check)
+        for key in headers:
+            if key.lower() == "traceparent":
+                return
+
+        # Inject traceparent - works as fallback when OTEL instrumentation doesn't cover this call
+        with contextlib.suppress(Exception):
+            from core.helper.trace_id_helper import generate_traceparent_header
+
+            traceparent = generate_traceparent_header()
+            if traceparent:
+                headers["traceparent"] = traceparent
+
     def _stream_request(
         self,
         method: str,

api/core/provider_manager.py

@@ -331,7 +331,6 @@ class ProviderManager:
                 provider=provider_schema.provider,
                 label=provider_schema.label,
                 icon_small=provider_schema.icon_small,
-                icon_large=provider_schema.icon_large,
                 supported_model_types=provider_schema.supported_model_types,
             ),
         )

api/core/rag/cleaner/clean_processor.py

@@ -27,26 +27,44 @@ class CleanProcessor:
                     pattern = r"([a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+)"
                     text = re.sub(pattern, "", text)
 
-                    # Remove URL but keep Markdown image URLs
-                    # First, temporarily replace Markdown image URLs with a placeholder
-                    markdown_image_pattern = r"!\[.*?\]\((https?://[^\s)]+)\)"
-                    placeholders: list[str] = []
+                    # Remove URL but keep Markdown image URLs and link URLs
+                    # Replace the ENTIRE markdown link/image with a single placeholder to protect
+                    # the link text (which might also be a URL) from being removed
+                    markdown_link_pattern = r"\[([^\]]*)\]\((https?://[^)]+)\)"
+                    markdown_image_pattern = r"!\[.*?\]\((https?://[^)]+)\)"
+                    placeholders: list[tuple[str, str, str]] = []  # (type, text, url)
 
-                    def replace_with_placeholder(match, placeholders=placeholders):
+                    def replace_markdown_with_placeholder(match, placeholders=placeholders):
+                        link_type = "link"
+                        link_text = match.group(1)
+                        url = match.group(2)
+                        placeholder = f"__MARKDOWN_PLACEHOLDER_{len(placeholders)}__"
+                        placeholders.append((link_type, link_text, url))
+                        return placeholder
+
+                    def replace_image_with_placeholder(match, placeholders=placeholders):
+                        link_type = "image"
                         url = match.group(1)
-                        placeholder = f"__MARKDOWN_IMAGE_URL_{len(placeholders)}__"
-                        placeholders.append(url)
-                        return f"![image]({placeholder})"
+                        placeholder = f"__MARKDOWN_PLACEHOLDER_{len(placeholders)}__"
+                        placeholders.append((link_type, "image", url))
+                        return placeholder
 
-                    text = re.sub(markdown_image_pattern, replace_with_placeholder, text)
+                    # Protect markdown links first
+                    text = re.sub(markdown_link_pattern, replace_markdown_with_placeholder, text)
+                    # Then protect markdown images
+                    text = re.sub(markdown_image_pattern, replace_image_with_placeholder, text)
 
                     # Now remove all remaining URLs
-                    url_pattern = r"https?://[^\s)]+"
+                    url_pattern = r"https?://\S+"
                     text = re.sub(url_pattern, "", text)
 
-                    # Finally, restore the Markdown image URLs
-                    for i, url in enumerate(placeholders):
-                        text = text.replace(f"__MARKDOWN_IMAGE_URL_{i}__", url)
+                    # Restore the Markdown links and images
+                    for i, (link_type, text_or_alt, url) in enumerate(placeholders):
+                        placeholder = f"__MARKDOWN_PLACEHOLDER_{i}__"
+                        if link_type == "link":
+                            text = text.replace(placeholder, f"[{text_or_alt}]({url})")
+                        else:  # image
+                            text = text.replace(placeholder, f"![{text_or_alt}]({url})")
         return text
 
     def filter_string(self, text):

api/core/rag/datasource/retrieval_service.py

@@ -1,4 +1,5 @@
 import concurrent.futures
+import logging
 from concurrent.futures import ThreadPoolExecutor
 from typing import Any
 
@@ -13,7 +14,7 @@ from core.model_runtime.entities.model_entities import ModelType
 from core.rag.data_post_processor.data_post_processor import DataPostProcessor
 from core.rag.datasource.keyword.keyword_factory import Keyword
 from core.rag.datasource.vdb.vector_factory import Vector
-from core.rag.embedding.retrieval import RetrievalSegments
+from core.rag.embedding.retrieval import RetrievalChildChunk, RetrievalSegments
 from core.rag.entities.metadata_entities import MetadataCondition
 from core.rag.index_processor.constant.doc_type import DocType
 from core.rag.index_processor.constant.index_type import IndexStructureType
@@ -36,6 +37,8 @@ default_retrieval_model = {
     "score_threshold_enabled": False,
 }
 
+logger = logging.getLogger(__name__)
+
 
 class RetrievalService:
     # Cache precompiled regular expressions to avoid repeated compilation
@@ -106,7 +109,12 @@ class RetrievalService:
                         )
                     )
 
-            concurrent.futures.wait(futures, timeout=3600, return_when=concurrent.futures.ALL_COMPLETED)
+            if futures:
+                for future in concurrent.futures.as_completed(futures, timeout=3600):
+                    if exceptions:
+                        for f in futures:
+                            f.cancel()
+                        break
 
         if exceptions:
             raise ValueError(";\n".join(exceptions))
@@ -210,6 +218,7 @@ class RetrievalService:
                 )
                 all_documents.extend(documents)
             except Exception as e:
+                logger.error(e, exc_info=True)
                 exceptions.append(str(e))
 
     @classmethod
@@ -303,6 +312,7 @@ class RetrievalService:
                     else:
                         all_documents.extend(documents)
             except Exception as e:
+                logger.error(e, exc_info=True)
                 exceptions.append(str(e))
 
     @classmethod
@@ -351,6 +361,7 @@ class RetrievalService:
                     else:
                         all_documents.extend(documents)
             except Exception as e:
+                logger.error(e, exc_info=True)
                 exceptions.append(str(e))
 
     @staticmethod
@@ -381,10 +392,9 @@ class RetrievalService:
             records = []
             include_segment_ids = set()
             segment_child_map = {}
-            segment_file_map = {}
 
             valid_dataset_documents = {}
-            image_doc_ids = []
+            image_doc_ids: list[Any] = []
             child_index_node_ids = []
             index_node_ids = []
             doc_to_document_map = {}
@@ -417,28 +427,39 @@ class RetrievalService:
             child_index_node_ids = [i for i in child_index_node_ids if i]
             index_node_ids = [i for i in index_node_ids if i]
 
-            segment_ids = []
+            segment_ids: list[str] = []
             index_node_segments: list[DocumentSegment] = []
             segments: list[DocumentSegment] = []
-            attachment_map = {}
-            child_chunk_map = {}
-            doc_segment_map = {}
+            attachment_map: dict[str, list[dict[str, Any]]] = {}
+            child_chunk_map: dict[str, list[ChildChunk]] = {}
+            doc_segment_map: dict[str, list[str]] = {}
 
             with session_factory.create_session() as session:
                 attachments = cls.get_segment_attachment_infos(image_doc_ids, session)
 
                 for attachment in attachments:
                     segment_ids.append(attachment["segment_id"])
-                    attachment_map[attachment["segment_id"]] = attachment
-                    doc_segment_map[attachment["segment_id"]] = attachment["attachment_id"]
-
+                    if attachment["segment_id"] in attachment_map:
+                        attachment_map[attachment["segment_id"]].append(attachment["attachment_info"])
+                    else:
+                        attachment_map[attachment["segment_id"]] = [attachment["attachment_info"]]
+                    if attachment["segment_id"] in doc_segment_map:
+                        doc_segment_map[attachment["segment_id"]].append(attachment["attachment_id"])
+                    else:
+                        doc_segment_map[attachment["segment_id"]] = [attachment["attachment_id"]]
                 child_chunk_stmt = select(ChildChunk).where(ChildChunk.index_node_id.in_(child_index_node_ids))
                 child_index_nodes = session.execute(child_chunk_stmt).scalars().all()
 
                 for i in child_index_nodes:
                     segment_ids.append(i.segment_id)
-                    child_chunk_map[i.segment_id] = i
-                    doc_segment_map[i.segment_id] = i.index_node_id
+                    if i.segment_id in child_chunk_map:
+                        child_chunk_map[i.segment_id].append(i)
+                    else:
+                        child_chunk_map[i.segment_id] = [i]
+                    if i.segment_id in doc_segment_map:
+                        doc_segment_map[i.segment_id].append(i.index_node_id)
+                    else:
+                        doc_segment_map[i.segment_id] = [i.index_node_id]
 
                 if index_node_ids:
                     document_segment_stmt = select(DocumentSegment).where(
@@ -448,7 +469,7 @@ class RetrievalService:
                     )
                     index_node_segments = session.execute(document_segment_stmt).scalars().all()  # type: ignore
                     for index_node_segment in index_node_segments:
-                        doc_segment_map[index_node_segment.id] = index_node_segment.index_node_id
+                        doc_segment_map[index_node_segment.id] = [index_node_segment.index_node_id]
                 if segment_ids:
                     document_segment_stmt = select(DocumentSegment).where(
                         DocumentSegment.enabled == True,
@@ -461,95 +482,86 @@ class RetrievalService:
                     segments.extend(index_node_segments)
 
             for segment in segments:
-                doc_id = doc_segment_map.get(segment.id)
-                child_chunk = child_chunk_map.get(segment.id)
-                attachment_info = attachment_map.get(segment.id)
+                child_chunks: list[ChildChunk] = child_chunk_map.get(segment.id, [])
+                attachment_infos: list[dict[str, Any]] = attachment_map.get(segment.id, [])
+                ds_dataset_document: DatasetDocument | None = valid_dataset_documents.get(segment.document_id)
 
-                if doc_id:
-                    document = doc_to_document_map[doc_id]
-                    ds_dataset_document: DatasetDocument | None = valid_dataset_documents.get(
-                        document.metadata.get("document_id")
-                    )
-
-                    if ds_dataset_document and ds_dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
-                        if segment.id not in include_segment_ids:
-                            include_segment_ids.add(segment.id)
-                            if child_chunk:
+                if ds_dataset_document and ds_dataset_document.doc_form == IndexStructureType.PARENT_CHILD_INDEX:
+                    if segment.id not in include_segment_ids:
+                        include_segment_ids.add(segment.id)
+                        if child_chunks or attachment_infos:
+                            child_chunk_details = []
+                            max_score = 0.0
+                            for child_chunk in child_chunks:
+                                document = doc_to_document_map[child_chunk.index_node_id]
                                 child_chunk_detail = {
                                     "id": child_chunk.id,
                                     "content": child_chunk.content,
                                     "position": child_chunk.position,
                                     "score": document.metadata.get("score", 0.0) if document else 0.0,
                                 }
-                                map_detail = {
-                                    "max_score": document.metadata.get("score", 0.0) if document else 0.0,
-                                    "child_chunks": [child_chunk_detail],
-                                }
-                                segment_child_map[segment.id] = map_detail
-                            record = {
-                                "segment": segment,
+                                child_chunk_details.append(child_chunk_detail)
+                                max_score = max(max_score, document.metadata.get("score", 0.0) if document else 0.0)
+                            for attachment_info in attachment_infos:
+                                file_document = doc_to_document_map[attachment_info["id"]]
+                                max_score = max(
+                                    max_score, file_document.metadata.get("score", 0.0) if file_document else 0.0
+                                )
+
+                            map_detail = {
+                                "max_score": max_score,
+                                "child_chunks": child_chunk_details,
                             }
-                            if attachment_info:
-                                segment_file_map[segment.id] = [attachment_info]
-                            records.append(record)
-                        else:
-                            if child_chunk:
-                                child_chunk_detail = {
-                                    "id": child_chunk.id,
-                                    "content": child_chunk.content,
-                                    "position": child_chunk.position,
-                                    "score": document.metadata.get("score", 0.0),
-                                }
-                                if segment.id in segment_child_map:
-                                    segment_child_map[segment.id]["child_chunks"].append(child_chunk_detail)  # type: ignore
-                                    segment_child_map[segment.id]["max_score"] = max(
-                                        segment_child_map[segment.id]["max_score"],
-                                        document.metadata.get("score", 0.0) if document else 0.0,
-                                    )
-                                else:
-                                    segment_child_map[segment.id] = {
-                                        "max_score": document.metadata.get("score", 0.0) if document else 0.0,
-                                        "child_chunks": [child_chunk_detail],
-                                    }
-                            if attachment_info:
-                                if segment.id in segment_file_map:
-                                    segment_file_map[segment.id].append(attachment_info)
-                                else:
-                                    segment_file_map[segment.id] = [attachment_info]
-                    else:
-                        if segment.id not in include_segment_ids:
-                            include_segment_ids.add(segment.id)
-                            record = {
-                                "segment": segment,
-                                "score": document.metadata.get("score", 0.0),  # type: ignore
-                            }
-                            if attachment_info:
-                                segment_file_map[segment.id] = [attachment_info]
-                            records.append(record)
-                        else:
-                            if attachment_info:
-                                attachment_infos = segment_file_map.get(segment.id, [])
-                                if attachment_info not in attachment_infos:
-                                    attachment_infos.append(attachment_info)
-                                segment_file_map[segment.id] = attachment_infos
+                            segment_child_map[segment.id] = map_detail
+                        record: dict[str, Any] = {
+                            "segment": segment,
+                        }
+                        records.append(record)
+                else:
+                    if segment.id not in include_segment_ids:
+                        include_segment_ids.add(segment.id)
+                        max_score = 0.0
+                        segment_document = doc_to_document_map.get(segment.index_node_id)
+                        if segment_document:
+                            max_score = max(max_score, segment_document.metadata.get("score", 0.0))
+                        for attachment_info in attachment_infos:
+                            file_doc = doc_to_document_map.get(attachment_info["id"])
+                            if file_doc:
+                                max_score = max(max_score, file_doc.metadata.get("score", 0.0))
+                        record = {
+                            "segment": segment,
+                            "score": max_score,
+                        }
+                        records.append(record)
 
             # Add child chunks information to records
             for record in records:
                 if record["segment"].id in segment_child_map:
                     record["child_chunks"] = segment_child_map[record["segment"].id].get("child_chunks")  # type: ignore
                     record["score"] = segment_child_map[record["segment"].id]["max_score"]  # type: ignore
-                if record["segment"].id in segment_file_map:
-                    record["files"] = segment_file_map[record["segment"].id]  # type: ignore[assignment]
+                if record["segment"].id in attachment_map:
+                    record["files"] = attachment_map[record["segment"].id]  # type: ignore[assignment]
 
-            result = []
+            result: list[RetrievalSegments] = []
             for record in records:
                 # Extract segment
                 segment = record["segment"]
 
                 # Extract child_chunks, ensuring it's a list or None
-                child_chunks = record.get("child_chunks")
-                if not isinstance(child_chunks, list):
-                    child_chunks = None
+                raw_child_chunks = record.get("child_chunks")
+                child_chunks_list: list[RetrievalChildChunk] | None = None
+                if isinstance(raw_child_chunks, list):
+                    # Sort by score descending
+                    sorted_chunks = sorted(raw_child_chunks, key=lambda x: x.get("score", 0.0), reverse=True)
+                    child_chunks_list = [
+                        RetrievalChildChunk(
+                            id=chunk["id"],
+                            content=chunk["content"],
+                            score=chunk.get("score", 0.0),
+                            position=chunk["position"],
+                        )
+                        for chunk in sorted_chunks
+                    ]
 
                 # Extract files, ensuring it's a list or None
                 files = record.get("files")
@@ -566,11 +578,11 @@ class RetrievalService:
 
                 # Create RetrievalSegments object
                 retrieval_segment = RetrievalSegments(
-                    segment=segment, child_chunks=child_chunks, score=score, files=files
+                    segment=segment, child_chunks=child_chunks_list, score=score, files=files
                 )
                 result.append(retrieval_segment)
 
-            return result
+            return sorted(result, key=lambda x: x.score if x.score is not None else 0.0, reverse=True)
         except Exception as e:
             db.session.rollback()
             raise e
@@ -662,7 +674,14 @@ class RetrievalService:
                             document_ids_filter=document_ids_filter,
                         )
                     )
-                concurrent.futures.wait(futures, timeout=300, return_when=concurrent.futures.ALL_COMPLETED)
+                # Use as_completed for early error propagation - cancel remaining futures on first error
+                if futures:
+                    for future in concurrent.futures.as_completed(futures, timeout=300):
+                        if future.exception():
+                            # Cancel remaining futures to avoid unnecessary waiting
+                            for f in futures:
+                                f.cancel()
+                            break
 
             if exceptions:
                 raise ValueError(";\n".join(exceptions))

api/core/rag/datasource/vdb/pgvector/pgvector.py

@@ -255,7 +255,10 @@ class PGVector(BaseVector):
                 return
 
             with self._get_cursor() as cur:
-                cur.execute("CREATE EXTENSION IF NOT EXISTS vector")
+                cur.execute("SELECT 1 FROM pg_extension WHERE extname = 'vector'")
+                if not cur.fetchone():
+                    cur.execute("CREATE EXTENSION vector")
+
                 cur.execute(SQL_CREATE_TABLE.format(table_name=self.table_name, dimension=dimension))
                 # PG hnsw index only support 2000 dimension or less
                 # ref: https://github.com/pgvector/pgvector?tab=readme-ov-file#indexing

api/core/rag/extractor/extract_processor.py

@@ -112,7 +112,7 @@ class ExtractProcessor:
                     if file_extension in {".xlsx", ".xls"}:
                         extractor = ExcelExtractor(file_path)
                     elif file_extension == ".pdf":
-                        extractor = PdfExtractor(file_path)
+                        extractor = PdfExtractor(file_path, upload_file.tenant_id, upload_file.created_by)
                     elif file_extension in {".md", ".markdown", ".mdx"}:
                         extractor = (
                             UnstructuredMarkdownExtractor(file_path, unstructured_api_url, unstructured_api_key)
@@ -148,7 +148,7 @@ class ExtractProcessor:
                     if file_extension in {".xlsx", ".xls"}:
                         extractor = ExcelExtractor(file_path)
                     elif file_extension == ".pdf":
-                        extractor = PdfExtractor(file_path)
+                        extractor = PdfExtractor(file_path, upload_file.tenant_id, upload_file.created_by)
                     elif file_extension in {".md", ".markdown", ".mdx"}:
                         extractor = MarkdownExtractor(file_path, autodetect_encoding=True)
                     elif file_extension in {".htm", ".html"}:

api/core/rag/extractor/pdf_extractor.py

@@ -1,25 +1,57 @@
 """Abstract interface for document loader implementations."""
 
 import contextlib
+import io
+import logging
+import uuid
 from collections.abc import Iterator
 
+import pypdfium2
+import pypdfium2.raw as pdfium_c
+
+from configs import dify_config
 from core.rag.extractor.blob.blob import Blob
 from core.rag.extractor.extractor_base import BaseExtractor
 from core.rag.models.document import Document
+from extensions.ext_database import db
 from extensions.ext_storage import storage
+from libs.datetime_utils import naive_utc_now
+from models.enums import CreatorUserRole
+from models.model import UploadFile
+
+logger = logging.getLogger(__name__)
 
 
 class PdfExtractor(BaseExtractor):
-    """Load pdf files.
-
+    """
+    PdfExtractor is used to extract text and images from PDF files.
 
     Args:
-        file_path: Path to the file to load.
+        file_path: Path to the PDF file.
+        tenant_id: Workspace ID.
+        user_id: ID of the user performing the extraction.
+        file_cache_key: Optional cache key for the extracted text.
     """
 
-    def __init__(self, file_path: str, file_cache_key: str | None = None):
-        """Initialize with file path."""
+    # Magic bytes for image format detection: (magic_bytes, extension, mime_type)
+    IMAGE_FORMATS = [
+        (b"\xff\xd8\xff", "jpg", "image/jpeg"),
+        (b"\x89PNG\r\n\x1a\n", "png", "image/png"),
+        (b"\x00\x00\x00\x0c\x6a\x50\x20\x20\x0d\x0a\x87\x0a", "jp2", "image/jp2"),
+        (b"GIF8", "gif", "image/gif"),
+        (b"BM", "bmp", "image/bmp"),
+        (b"II*\x00", "tiff", "image/tiff"),
+        (b"MM\x00*", "tiff", "image/tiff"),
+        (b"II+\x00", "tiff", "image/tiff"),
+        (b"MM\x00+", "tiff", "image/tiff"),
+    ]
+    MAX_MAGIC_LEN = max(len(m) for m, _, _ in IMAGE_FORMATS)
+
+    def __init__(self, file_path: str, tenant_id: str, user_id: str, file_cache_key: str | None = None):
+        """Initialize PdfExtractor."""
         self._file_path = file_path
+        self._tenant_id = tenant_id
+        self._user_id = user_id
         self._file_cache_key = file_cache_key
 
     def extract(self) -> list[Document]:
@@ -50,7 +82,6 @@ class PdfExtractor(BaseExtractor):
 
     def parse(self, blob: Blob) -> Iterator[Document]:
         """Lazily parse the blob."""
-        import pypdfium2  # type: ignore
 
         with blob.as_bytes_io() as file_path:
             pdf_reader = pypdfium2.PdfDocument(file_path, autoclose=True)
@@ -59,8 +90,87 @@ class PdfExtractor(BaseExtractor):
                     text_page = page.get_textpage()
                     content = text_page.get_text_range()
                     text_page.close()
+
+                    image_content = self._extract_images(page)
+                    if image_content:
+                        content += "\n" + image_content
+
                     page.close()
                     metadata = {"source": blob.source, "page": page_number}
                     yield Document(page_content=content, metadata=metadata)
             finally:
                 pdf_reader.close()
+
+    def _extract_images(self, page) -> str:
+        """
+        Extract images from a PDF page, save them to storage and database,
+        and return markdown image links.
+
+        Args:
+            page: pypdfium2 page object.
+
+        Returns:
+            Markdown string containing links to the extracted images.
+        """
+        image_content = []
+        upload_files = []
+        base_url = dify_config.INTERNAL_FILES_URL or dify_config.FILES_URL
+
+        try:
+            image_objects = page.get_objects(filter=(pdfium_c.FPDF_PAGEOBJ_IMAGE,))
+            for obj in image_objects:
+                try:
+                    # Extract image bytes
+                    img_byte_arr = io.BytesIO()
+                    # Extract DCTDecode (JPEG) and JPXDecode (JPEG 2000) images directly
+                    # Fallback to png for other formats
+                    obj.extract(img_byte_arr, fb_format="png")
+                    img_bytes = img_byte_arr.getvalue()
+
+                    if not img_bytes:
+                        continue
+
+                    header = img_bytes[: self.MAX_MAGIC_LEN]
+                    image_ext = None
+                    mime_type = None
+                    for magic, ext, mime in self.IMAGE_FORMATS:
+                        if header.startswith(magic):
+                            image_ext = ext
+                            mime_type = mime
+                            break
+
+                    if not image_ext or not mime_type:
+                        continue
+
+                    file_uuid = str(uuid.uuid4())
+                    file_key = "image_files/" + self._tenant_id + "/" + file_uuid + "." + image_ext
+
+                    storage.save(file_key, img_bytes)
+
+                    # save file to db
+                    upload_file = UploadFile(
+                        tenant_id=self._tenant_id,
+                        storage_type=dify_config.STORAGE_TYPE,
+                        key=file_key,
+                        name=file_key,
+                        size=len(img_bytes),
+                        extension=image_ext,
+                        mime_type=mime_type,
+                        created_by=self._user_id,
+                        created_by_role=CreatorUserRole.ACCOUNT,
+                        created_at=naive_utc_now(),
+                        used=True,
+                        used_by=self._user_id,
+                        used_at=naive_utc_now(),
+                    )
+                    upload_files.append(upload_file)
+                    image_content.append(f"![image]({base_url}/files/{upload_file.id}/file-preview)")
+                except Exception as e:
+                    logger.warning("Failed to extract image from PDF: %s", e)
+                    continue
+        except Exception as e:
+            logger.warning("Failed to get objects from PDF page: %s", e)
+        if upload_files:
+            db.session.add_all(upload_files)
+            db.session.commit()
+        return "\n".join(image_content)

api/core/rag/retrieval/dataset_retrieval.py

@@ -7,7 +7,7 @@ from collections.abc import Generator, Mapping
 from typing import Any, Union, cast
 
 from flask import Flask, current_app
-from sqlalchemy import and_, or_, select
+from sqlalchemy import and_, literal, or_, select
 from sqlalchemy.orm import Session
 
 from core.app.app_config.entities import (
@@ -516,6 +516,9 @@ class DatasetRetrieval:
                     ].embedding_model_provider
                     weights["vector_setting"]["embedding_model_name"] = available_datasets[0].embedding_model
         with measure_time() as timer:
+            cancel_event = threading.Event()
+            thread_exceptions: list[Exception] = []
+
             if query:
                 query_thread = threading.Thread(
                     target=self._multiple_retrieve_thread,
@@ -534,6 +537,8 @@ class DatasetRetrieval:
                         "score_threshold": score_threshold,
                         "query": query,
                         "attachment_id": None,
+                        "cancel_event": cancel_event,
+                        "thread_exceptions": thread_exceptions,
                     },
                 )
                 all_threads.append(query_thread)
@@ -557,12 +562,25 @@ class DatasetRetrieval:
                             "score_threshold": score_threshold,
                             "query": None,
                             "attachment_id": attachment_id,
+                            "cancel_event": cancel_event,
+                            "thread_exceptions": thread_exceptions,
                         },
                     )
                     all_threads.append(attachment_thread)
                     attachment_thread.start()
-            for thread in all_threads:
-                thread.join()
+
+            # Poll threads with short timeout to detect errors quickly (fail-fast)
+            while any(t.is_alive() for t in all_threads):
+                for thread in all_threads:
+                    thread.join(timeout=0.1)
+                    if thread_exceptions:
+                        cancel_event.set()
+                        break
+                if thread_exceptions:
+                    break
+
+            if thread_exceptions:
+                raise thread_exceptions[0]
         self._on_query(query, attachment_ids, dataset_ids, app_id, user_from, user_id)
 
         if all_documents:
@@ -1036,7 +1054,7 @@ class DatasetRetrieval:
             if automatic_metadata_filters:
                 conditions = []
                 for sequence, filter in enumerate(automatic_metadata_filters):
-                    self._process_metadata_filter_func(
+                    self.process_metadata_filter_func(
                         sequence,
                         filter.get("condition"),  # type: ignore
                         filter.get("metadata_name"),  # type: ignore
@@ -1072,7 +1090,7 @@ class DatasetRetrieval:
                             value=expected_value,
                         )
                     )
-                    filters = self._process_metadata_filter_func(
+                    filters = self.process_metadata_filter_func(
                         sequence,
                         condition.comparison_operator,
                         metadata_name,
@@ -1168,8 +1186,9 @@ class DatasetRetrieval:
             return None
         return automatic_metadata_filters
 
-    def _process_metadata_filter_func(
-        self, sequence: int, condition: str, metadata_name: str, value: Any | None, filters: list
+    @classmethod
+    def process_metadata_filter_func(
+        cls, sequence: int, condition: str, metadata_name: str, value: Any | None, filters: list
     ):
         if value is None and condition not in ("empty", "not empty"):
             return filters
@@ -1218,6 +1237,20 @@ class DatasetRetrieval:
 
             case "≥" | ">=":
                 filters.append(DatasetDocument.doc_metadata[metadata_name].as_float() >= value)
+            case "in" | "not in":
+                if isinstance(value, str):
+                    value_list = [v.strip() for v in value.split(",") if v.strip()]
+                elif isinstance(value, (list, tuple)):
+                    value_list = [str(v) for v in value if v is not None]
+                else:
+                    value_list = [str(value)] if value is not None else []
+
+                if not value_list:
+                    # `field in []` is False, `field not in []` is True
+                    filters.append(literal(condition == "not in"))
+                else:
+                    op = json_field.in_ if condition == "in" else json_field.notin_
+                    filters.append(op(value_list))
             case _:
                 pass
 
@@ -1389,40 +1422,53 @@ class DatasetRetrieval:
         score_threshold: float,
         query: str | None,
         attachment_id: str | None,
+        cancel_event: threading.Event | None = None,
+        thread_exceptions: list[Exception] | None = None,
     ):
-        with flask_app.app_context():
-            threads = []
-            all_documents_item: list[Document] = []
-            index_type = None
-            for dataset in available_datasets:
-                index_type = dataset.indexing_technique
-                document_ids_filter = None
-                if dataset.provider != "external":
-                    if metadata_condition and not metadata_filter_document_ids:
-                        continue
-                    if metadata_filter_document_ids:
-                        document_ids = metadata_filter_document_ids.get(dataset.id, [])
-                        if document_ids:
-                            document_ids_filter = document_ids
-                        else:
+        try:
+            with flask_app.app_context():
+                threads = []
+                all_documents_item: list[Document] = []
+                index_type = None
+                for dataset in available_datasets:
+                    # Check for cancellation signal
+                    if cancel_event and cancel_event.is_set():
+                        break
+                    index_type = dataset.indexing_technique
+                    document_ids_filter = None
+                    if dataset.provider != "external":
+                        if metadata_condition and not metadata_filter_document_ids:
                             continue
-                retrieval_thread = threading.Thread(
-                    target=self._retriever,
-                    kwargs={
-                        "flask_app": flask_app,
-                        "dataset_id": dataset.id,
-                        "query": query,
-                        "top_k": top_k,
-                        "all_documents": all_documents_item,
-                        "document_ids_filter": document_ids_filter,
-                        "metadata_condition": metadata_condition,
-                        "attachment_ids": [attachment_id] if attachment_id else None,
-                    },
-                )
-                threads.append(retrieval_thread)
-                retrieval_thread.start()
-            for thread in threads:
-                thread.join()
+                        if metadata_filter_document_ids:
+                            document_ids = metadata_filter_document_ids.get(dataset.id, [])
+                            if document_ids:
+                                document_ids_filter = document_ids
+                            else:
+                                continue
+                    retrieval_thread = threading.Thread(
+                        target=self._retriever,
+                        kwargs={
+                            "flask_app": flask_app,
+                            "dataset_id": dataset.id,
+                            "query": query,
+                            "top_k": top_k,
+                            "all_documents": all_documents_item,
+                            "document_ids_filter": document_ids_filter,
+                            "metadata_condition": metadata_condition,
+                            "attachment_ids": [attachment_id] if attachment_id else None,
+                        },
+                    )
+                    threads.append(retrieval_thread)
+                    retrieval_thread.start()
+
+                # Poll threads with short timeout to respond quickly to cancellation
+                while any(t.is_alive() for t in threads):
+                    for thread in threads:
+                        thread.join(timeout=0.1)
+                        if cancel_event and cancel_event.is_set():
+                            break
+                    if cancel_event and cancel_event.is_set():
+                        break
 
             if reranking_enable:
                 # do rerank for searched documents
@@ -1455,3 +1501,8 @@ class DatasetRetrieval:
                     all_documents_item = all_documents_item[:top_k] if top_k else all_documents_item
             if all_documents_item:
                 all_documents.extend(all_documents_item)
+        except Exception as e:
+            if cancel_event:
+                cancel_event.set()
+            if thread_exceptions is not None:
+                thread_exceptions.append(e)

api/core/tools/mcp_tool/tool.py

@@ -6,7 +6,15 @@ from typing import Any
 
 from core.mcp.auth_client import MCPClientWithAuthRetry
 from core.mcp.error import MCPConnectionError
-from core.mcp.types import AudioContent, CallToolResult, ImageContent, TextContent
+from core.mcp.types import (
+    AudioContent,
+    BlobResourceContents,
+    CallToolResult,
+    EmbeddedResource,
+    ImageContent,
+    TextContent,
+    TextResourceContents,
+)
 from core.tools.__base.tool import Tool
 from core.tools.__base.tool_runtime import ToolRuntime
 from core.tools.entities.tool_entities import ToolEntity, ToolInvokeMessage, ToolProviderType
@@ -53,10 +61,19 @@ class MCPTool(Tool):
         for content in result.content:
             if isinstance(content, TextContent):
                 yield from self._process_text_content(content)
-            elif isinstance(content, ImageContent):
-                yield self._process_image_content(content)
-            elif isinstance(content, AudioContent):
-                yield self._process_audio_content(content)
+            elif isinstance(content, ImageContent | AudioContent):
+                yield self.create_blob_message(
+                    blob=base64.b64decode(content.data), meta={"mime_type": content.mimeType}
+                )
+            elif isinstance(content, EmbeddedResource):
+                resource = content.resource
+                if isinstance(resource, TextResourceContents):
+                    yield self.create_text_message(resource.text)
+                elif isinstance(resource, BlobResourceContents):
+                    mime_type = resource.mimeType or "application/octet-stream"
+                    yield self.create_blob_message(blob=base64.b64decode(resource.blob), meta={"mime_type": mime_type})
+                else:
+                    raise ToolInvokeError(f"Unsupported embedded resource type: {type(resource)}")
             else:
                 logger.warning("Unsupported content type=%s", type(content))
 
@@ -101,14 +118,6 @@ class MCPTool(Tool):
         for item in json_list:
             yield self.create_json_message(item)
 
-    def _process_image_content(self, content: ImageContent) -> ToolInvokeMessage:
-        """Process image content and return a blob message."""
-        return self.create_blob_message(blob=base64.b64decode(content.data), meta={"mime_type": content.mimeType})
-
-    def _process_audio_content(self, content: AudioContent) -> ToolInvokeMessage:
-        """Process audio content and return a blob message."""
-        return self.create_blob_message(blob=base64.b64decode(content.data), meta={"mime_type": content.mimeType})
-
     def fork_tool_runtime(self, runtime: ToolRuntime) -> "MCPTool":
         return MCPTool(
             entity=self.entity,

api/core/tools/utils/parser.py

@@ -378,7 +378,7 @@ class ApiBasedToolSchemaParser:
     @staticmethod
     def auto_parse_to_tool_bundle(
         content: str, extra_info: dict | None = None, warning: dict | None = None
-    ) -> tuple[list[ApiToolBundle], str]:
+    ) -> tuple[list[ApiToolBundle], ApiProviderSchemaType]:
         """
         auto parse to tool bundle
 

api/core/tools/utils/text_processing_utils.py

@@ -4,6 +4,7 @@ import re
 def remove_leading_symbols(text: str) -> str:
     """
     Remove leading punctuation or symbols from the given text.
+    Preserves markdown links like [text](url) at the start.
 
     Args:
         text (str): The input text to process.
@@ -11,6 +12,11 @@ def remove_leading_symbols(text: str) -> str:
     Returns:
         str: The text with leading punctuation or symbols removed.
     """
+    # Check if text starts with a markdown link - preserve it
+    markdown_link_pattern = r"^\[([^\]]+)\]\((https?://[^)]+)\)"
+    if re.match(markdown_link_pattern, text):
+        return text
+
     # Match Unicode ranges for punctuation and symbols
     # FIXME this pattern is confused quick fix for #11868 maybe refactor it later
     pattern = r'^[\[\]\u2000-\u2025\u2027-\u206F\u2E00-\u2E7F\u3000-\u300F\u3011-\u303F"#$%&\'()*+,./:;<=>?@^_`~]+'

api/core/tools/workflow_as_tool/provider.py

@@ -5,6 +5,7 @@ from sqlalchemy.orm import Session
 
 from core.app.app_config.entities import VariableEntity, VariableEntityType
 from core.app.apps.workflow.app_config_manager import WorkflowAppConfigManager
+from core.db.session_factory import session_factory
 from core.plugin.entities.parameters import PluginParameterOption
 from core.tools.__base.tool_provider import ToolProviderController
 from core.tools.__base.tool_runtime import ToolRuntime
@@ -47,33 +48,29 @@ class WorkflowToolProviderController(ToolProviderController):
 
     @classmethod
     def from_db(cls, db_provider: WorkflowToolProvider) -> "WorkflowToolProviderController":
-        with Session(db.engine, expire_on_commit=False) as session, session.begin():
-            provider = session.get(WorkflowToolProvider, db_provider.id) if db_provider.id else None
-            if not provider:
-                raise ValueError("workflow provider not found")
-            app = session.get(App, provider.app_id)
+        with session_factory.create_session() as session, session.begin():
+            app = session.get(App, db_provider.app_id)
             if not app:
                 raise ValueError("app not found")
 
-            user = session.get(Account, provider.user_id) if provider.user_id else None
-
+            user = session.get(Account, db_provider.user_id) if db_provider.user_id else None
             controller = WorkflowToolProviderController(
                 entity=ToolProviderEntity(
                     identity=ToolProviderIdentity(
                         author=user.name if user else "",
-                        name=provider.label,
-                        label=I18nObject(en_US=provider.label, zh_Hans=provider.label),
-                        description=I18nObject(en_US=provider.description, zh_Hans=provider.description),
-                        icon=provider.icon,
+                        name=db_provider.label,
+                        label=I18nObject(en_US=db_provider.label, zh_Hans=db_provider.label),
+                        description=I18nObject(en_US=db_provider.description, zh_Hans=db_provider.description),
+                        icon=db_provider.icon,
                     ),
                     credentials_schema=[],
                     plugin_id=None,
                 ),
-                provider_id=provider.id or "",
+                provider_id=db_provider.id,
             )
 
             controller.tools = [
-                controller._get_db_provider_tool(provider, app, session=session, user=user),
+                controller._get_db_provider_tool(db_provider, app, session=session, user=user),
             ]
 
         return controller

api/core/workflow/graph_engine/graph_traversal/skip_propagator.py

@@ -60,6 +60,7 @@ class SkipPropagator:
         if edge_states["has_taken"]:
             # Enqueue node
             self._state_manager.enqueue_node(downstream_node_id)
+            self._state_manager.start_execution(downstream_node_id)
             return
 
         # All edges are skipped, propagate skip to this node

api/core/workflow/nodes/agent/exc.py

@@ -119,3 +119,14 @@ class AgentVariableTypeError(AgentNodeError):
         self.expected_type = expected_type
         self.actual_type = actual_type
         super().__init__(message)
+
+
+class AgentMaxIterationError(AgentNodeError):
+    """Exception raised when the agent exceeds the maximum iteration limit."""
+
+    def __init__(self, max_iteration: int):
+        self.max_iteration = max_iteration
+        super().__init__(
+            f"Agent exceeded the maximum iteration limit of {max_iteration}. "
+            f"The agent was unable to complete the task within the allowed number of iterations."
+        )

api/core/workflow/nodes/knowledge_retrieval/knowledge_retrieval_node.py

@@ -6,7 +6,7 @@ from collections import defaultdict
 from collections.abc import Mapping, Sequence
 from typing import TYPE_CHECKING, Any, cast
 
-from sqlalchemy import and_, func, literal, or_, select
+from sqlalchemy import and_, func, or_, select
 from sqlalchemy.orm import sessionmaker
 
 from core.app.app_config.entities import DatasetRetrieveConfigEntity
@@ -460,7 +460,7 @@ class KnowledgeRetrievalNode(LLMUsageTrackingMixin, Node[KnowledgeRetrievalNodeD
             if automatic_metadata_filters:
                 conditions = []
                 for sequence, filter in enumerate(automatic_metadata_filters):
-                    self._process_metadata_filter_func(
+                    DatasetRetrieval.process_metadata_filter_func(
                         sequence,
                         filter.get("condition", ""),
                         filter.get("metadata_name", ""),
@@ -504,7 +504,7 @@ class KnowledgeRetrievalNode(LLMUsageTrackingMixin, Node[KnowledgeRetrievalNodeD
                             value=expected_value,
                         )
                     )
-                    filters = self._process_metadata_filter_func(
+                    filters = DatasetRetrieval.process_metadata_filter_func(
                         sequence,
                         condition.comparison_operator,
                         metadata_name,
@@ -603,87 +603,6 @@ class KnowledgeRetrievalNode(LLMUsageTrackingMixin, Node[KnowledgeRetrievalNodeD
             return [], usage
         return automatic_metadata_filters, usage
 
-    def _process_metadata_filter_func(
-        self, sequence: int, condition: str, metadata_name: str, value: Any, filters: list[Any]
-    ) -> list[Any]:
-        if value is None and condition not in ("empty", "not empty"):
-            return filters
-
-        json_field = Document.doc_metadata[metadata_name].as_string()
-
-        match condition:
-            case "contains":
-                filters.append(json_field.like(f"%{value}%"))
-
-            case "not contains":
-                filters.append(json_field.notlike(f"%{value}%"))
-
-            case "start with":
-                filters.append(json_field.like(f"{value}%"))
-
-            case "end with":
-                filters.append(json_field.like(f"%{value}"))
-            case "in":
-                if isinstance(value, str):
-                    value_list = [v.strip() for v in value.split(",") if v.strip()]
-                elif isinstance(value, (list, tuple)):
-                    value_list = [str(v) for v in value if v is not None]
-                else:
-                    value_list = [str(value)] if value is not None else []
-
-                if not value_list:
-                    filters.append(literal(False))
-                else:
-                    filters.append(json_field.in_(value_list))
-
-            case "not in":
-                if isinstance(value, str):
-                    value_list = [v.strip() for v in value.split(",") if v.strip()]
-                elif isinstance(value, (list, tuple)):
-                    value_list = [str(v) for v in value if v is not None]
-                else:
-                    value_list = [str(value)] if value is not None else []
-
-                if not value_list:
-                    filters.append(literal(True))
-                else:
-                    filters.append(json_field.notin_(value_list))
-
-            case "is" | "=":
-                if isinstance(value, str):
-                    filters.append(json_field == value)
-                elif isinstance(value, (int, float)):
-                    filters.append(Document.doc_metadata[metadata_name].as_float() == value)
-
-            case "is not" | "≠":
-                if isinstance(value, str):
-                    filters.append(json_field != value)
-                elif isinstance(value, (int, float)):
-                    filters.append(Document.doc_metadata[metadata_name].as_float() != value)
-
-            case "empty":
-                filters.append(Document.doc_metadata[metadata_name].is_(None))
-
-            case "not empty":
-                filters.append(Document.doc_metadata[metadata_name].isnot(None))
-
-            case "before" | "<":
-                filters.append(Document.doc_metadata[metadata_name].as_float() < value)
-
-            case "after" | ">":
-                filters.append(Document.doc_metadata[metadata_name].as_float() > value)
-
-            case "≤" | "<=":
-                filters.append(Document.doc_metadata[metadata_name].as_float() <= value)
-
-            case "≥" | ">=":
-                filters.append(Document.doc_metadata[metadata_name].as_float() >= value)
-
-            case _:
-                pass
-
-        return filters
-
     @classmethod
     def _extract_variable_selector_to_variable_mapping(
         cls,

api/extensions/ext_celery.py

@@ -12,9 +12,8 @@ from dify_app import DifyApp
 
 def _get_celery_ssl_options() -> dict[str, Any] | None:
     """Get SSL configuration for Celery broker/backend connections."""
-    # Use REDIS_USE_SSL for consistency with the main Redis client
     # Only apply SSL if we're using Redis as broker/backend
-    if not dify_config.REDIS_USE_SSL:
+    if not dify_config.BROKER_USE_SSL:
         return None
 
     # Check if Celery is actually using Redis
@@ -47,7 +46,11 @@ def _get_celery_ssl_options() -> dict[str, Any] | None:
 def init_app(app: DifyApp) -> Celery:
     class FlaskTask(Task):
         def __call__(self, *args: object, **kwargs: object) -> object:
+            from core.logging.context import init_request_context
+
             with app.app_context():
+                # Initialize logging context for this task (similar to before_request in Flask)
+                init_request_context()
                 return self.run(*args, **kwargs)
 
     broker_transport_options = {}

api/extensions/ext_logging.py

@@ -1,18 +1,19 @@
+"""Logging extension for Dify Flask application."""
+
 import logging
 import os
 import sys
-import uuid
 from logging.handlers import RotatingFileHandler
 
-import flask
-
 from configs import dify_config
-from core.helper.trace_id_helper import get_trace_id_from_otel_context
 from dify_app import DifyApp
 
 
 def init_app(app: DifyApp):
+    """Initialize logging with support for text or JSON format."""
     log_handlers: list[logging.Handler] = []
+
+    # File handler
     log_file = dify_config.LOG_FILE
     if log_file:
         log_dir = os.path.dirname(log_file)
@@ -25,27 +26,53 @@ def init_app(app: DifyApp):
             )
         )
 
-    # Always add StreamHandler to log to console
+    # Console handler
     sh = logging.StreamHandler(sys.stdout)
     log_handlers.append(sh)
 
-    # Apply RequestIdFilter to all handlers
-    for handler in log_handlers:
-        handler.addFilter(RequestIdFilter())
+    # Apply filters to all handlers
+    from core.logging.filters import IdentityContextFilter, TraceContextFilter
 
+    for handler in log_handlers:
+        handler.addFilter(TraceContextFilter())
+        handler.addFilter(IdentityContextFilter())
+
+    # Configure formatter based on format type
+    formatter = _create_formatter()
+    for handler in log_handlers:
+        handler.setFormatter(formatter)
+
+    # Configure root logger
     logging.basicConfig(
         level=dify_config.LOG_LEVEL,
-        format=dify_config.LOG_FORMAT,
-        datefmt=dify_config.LOG_DATEFORMAT,
         handlers=log_handlers,
         force=True,
     )
 
-    # Apply RequestIdFormatter to all handlers
-    apply_request_id_formatter()
-
     # Disable propagation for noisy loggers to avoid duplicate logs
     logging.getLogger("sqlalchemy.engine").propagate = False
+
+    # Apply timezone if specified (only for text format)
+    if dify_config.LOG_OUTPUT_FORMAT == "text":
+        _apply_timezone(log_handlers)
+
+
+def _create_formatter() -> logging.Formatter:
+    """Create appropriate formatter based on configuration."""
+    if dify_config.LOG_OUTPUT_FORMAT == "json":
+        from core.logging.structured_formatter import StructuredJSONFormatter
+
+        return StructuredJSONFormatter()
+    else:
+        # Text format - use existing pattern with backward compatible formatter
+        return _TextFormatter(
+            fmt=dify_config.LOG_FORMAT,
+            datefmt=dify_config.LOG_DATEFORMAT,
+        )
+
+
+def _apply_timezone(handlers: list[logging.Handler]):
+    """Apply timezone conversion to text formatters."""
     log_tz = dify_config.LOG_TZ
     if log_tz:
         from datetime import datetime
@@ -57,34 +84,51 @@ def init_app(app: DifyApp):
         def time_converter(seconds):
             return datetime.fromtimestamp(seconds, tz=timezone).timetuple()
 
-        for handler in logging.root.handlers:
+        for handler in handlers:
             if handler.formatter:
-                handler.formatter.converter = time_converter
+                handler.formatter.converter = time_converter  # type: ignore[attr-defined]
 
 
-def get_request_id():
-    if getattr(flask.g, "request_id", None):
-        return flask.g.request_id
+class _TextFormatter(logging.Formatter):
+    """Text formatter that ensures trace_id and req_id are always present."""
 
-    new_uuid = uuid.uuid4().hex[:10]
-    flask.g.request_id = new_uuid
-
-    return new_uuid
+    def format(self, record: logging.LogRecord) -> str:
+        if not hasattr(record, "req_id"):
+            record.req_id = ""
+        if not hasattr(record, "trace_id"):
+            record.trace_id = ""
+        if not hasattr(record, "span_id"):
+            record.span_id = ""
+        return super().format(record)
 
 
+def get_request_id() -> str:
+    """Get request ID for current request context.
+
+    Deprecated: Use core.logging.context.get_request_id() directly.
+    """
+    from core.logging.context import get_request_id as _get_request_id
+
+    return _get_request_id()
+
+
+# Backward compatibility aliases
 class RequestIdFilter(logging.Filter):
-    # This is a logging filter that makes the request ID available for use in
-    # the logging format. Note that we're checking if we're in a request
-    # context, as we may want to log things before Flask is fully loaded.
-    def filter(self, record):
-        trace_id = get_trace_id_from_otel_context() or ""
-        record.req_id = get_request_id() if flask.has_request_context() else ""
-        record.trace_id = trace_id
+    """Deprecated: Use TraceContextFilter from core.logging.filters instead."""
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        from core.logging.context import get_request_id as _get_request_id
+        from core.logging.context import get_trace_id as _get_trace_id
+
+        record.req_id = _get_request_id()
+        record.trace_id = _get_trace_id()
         return True
 
 
 class RequestIdFormatter(logging.Formatter):
-    def format(self, record):
+    """Deprecated: Use _TextFormatter instead."""
+
+    def format(self, record: logging.LogRecord) -> str:
         if not hasattr(record, "req_id"):
             record.req_id = ""
         if not hasattr(record, "trace_id"):
@@ -93,6 +137,7 @@ class RequestIdFormatter(logging.Formatter):
 
 
 def apply_request_id_formatter():
+    """Deprecated: Formatter is now applied in init_app."""
     for handler in logging.root.handlers:
         if handler.formatter:
             handler.formatter = RequestIdFormatter(dify_config.LOG_FORMAT, dify_config.LOG_DATEFORMAT)

api/extensions/otel/instrumentation.py

@@ -19,26 +19,43 @@ logger = logging.getLogger(__name__)
 
 
 class ExceptionLoggingHandler(logging.Handler):
+    """
+    Handler that records exceptions to the current OpenTelemetry span.
+
+    Unlike creating a new span, this records exceptions on the existing span
+    to maintain trace context consistency throughout the request lifecycle.
+    """
+
     def emit(self, record: logging.LogRecord):
         with contextlib.suppress(Exception):
-            if record.exc_info:
-                tracer = get_tracer_provider().get_tracer("dify.exception.logging")
-                with tracer.start_as_current_span(
-                    "log.exception",
-                    attributes={
-                        "log.level": record.levelname,
-                        "log.message": record.getMessage(),
-                        "log.logger": record.name,
-                        "log.file.path": record.pathname,
-                        "log.file.line": record.lineno,
-                    },
-                ) as span:
-                    span.set_status(StatusCode.ERROR)
-                    if record.exc_info[1]:
-                        span.record_exception(record.exc_info[1])
-                        span.set_attribute("exception.message", str(record.exc_info[1]))
-                    if record.exc_info[0]:
-                        span.set_attribute("exception.type", record.exc_info[0].__name__)
+            if not record.exc_info:
+                return
+
+            from opentelemetry.trace import get_current_span
+
+            span = get_current_span()
+            if not span or not span.is_recording():
+                return
+
+            # Record exception on the current span instead of creating a new one
+            span.set_status(StatusCode.ERROR, record.getMessage())
+
+            # Add log context as span events/attributes
+            span.add_event(
+                "log.exception",
+                attributes={
+                    "log.level": record.levelname,
+                    "log.message": record.getMessage(),
+                    "log.logger": record.name,
+                    "log.file.path": record.pathname,
+                    "log.file.line": record.lineno,
+                },
+            )
+
+            if record.exc_info[1]:
+                span.record_exception(record.exc_info[1])
+            if record.exc_info[0]:
+                span.set_attribute("exception.type", record.exc_info[0].__name__)
 
 
 def instrument_exception_logging() -> None:

api/extensions/storage/tencent_cos_storage.py

@@ -13,12 +13,20 @@ class TencentCosStorage(BaseStorage):
         super().__init__()
 
         self.bucket_name = dify_config.TENCENT_COS_BUCKET_NAME
-        config = CosConfig(
-            Region=dify_config.TENCENT_COS_REGION,
-            SecretId=dify_config.TENCENT_COS_SECRET_ID,
-            SecretKey=dify_config.TENCENT_COS_SECRET_KEY,
-            Scheme=dify_config.TENCENT_COS_SCHEME,
-        )
+        if dify_config.TENCENT_COS_CUSTOM_DOMAIN:
+            config = CosConfig(
+                Domain=dify_config.TENCENT_COS_CUSTOM_DOMAIN,
+                SecretId=dify_config.TENCENT_COS_SECRET_ID,
+                SecretKey=dify_config.TENCENT_COS_SECRET_KEY,
+                Scheme=dify_config.TENCENT_COS_SCHEME,
+            )
+        else:
+            config = CosConfig(
+                Region=dify_config.TENCENT_COS_REGION,
+                SecretId=dify_config.TENCENT_COS_SECRET_ID,
+                SecretKey=dify_config.TENCENT_COS_SECRET_KEY,
+                Scheme=dify_config.TENCENT_COS_SCHEME,
+            )
         self.client = CosS3Client(config)
 
     def save(self, filename, data):

api/fields/annotation_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from libs.helper import TimestampField
 
@@ -12,7 +12,7 @@ annotation_fields = {
 }
 
 
-def build_annotation_model(api_or_ns: Api | Namespace):
+def build_annotation_model(api_or_ns: Namespace):
     """Build the annotation model for the API or Namespace."""
     return api_or_ns.model("Annotation", annotation_fields)
 

api/fields/conversation_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from fields.member_fields import simple_account_fields
 from libs.helper import TimestampField
@@ -46,7 +46,7 @@ message_file_fields = {
 }
 
 
-def build_message_file_model(api_or_ns: Api | Namespace):
+def build_message_file_model(api_or_ns: Namespace):
     """Build the message file fields for the API or Namespace."""
     return api_or_ns.model("MessageFile", message_file_fields)
 
@@ -217,7 +217,7 @@ conversation_infinite_scroll_pagination_fields = {
 }
 
 
-def build_conversation_infinite_scroll_pagination_model(api_or_ns: Api | Namespace):
+def build_conversation_infinite_scroll_pagination_model(api_or_ns: Namespace):
     """Build the conversation infinite scroll pagination model for the API or Namespace."""
     simple_conversation_model = build_simple_conversation_model(api_or_ns)
 
@@ -226,11 +226,11 @@ def build_conversation_infinite_scroll_pagination_model(api_or_ns: Api | Namespa
     return api_or_ns.model("ConversationInfiniteScrollPagination", copied_fields)
 
 
-def build_conversation_delete_model(api_or_ns: Api | Namespace):
+def build_conversation_delete_model(api_or_ns: Namespace):
     """Build the conversation delete model for the API or Namespace."""
     return api_or_ns.model("ConversationDelete", conversation_delete_fields)
 
 
-def build_simple_conversation_model(api_or_ns: Api | Namespace):
+def build_simple_conversation_model(api_or_ns: Namespace):
     """Build the simple conversation model for the API or Namespace."""
     return api_or_ns.model("SimpleConversation", simple_conversation_fields)

api/fields/conversation_variable_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from libs.helper import TimestampField
 
@@ -29,12 +29,12 @@ conversation_variable_infinite_scroll_pagination_fields = {
 }
 
 
-def build_conversation_variable_model(api_or_ns: Api | Namespace):
+def build_conversation_variable_model(api_or_ns: Namespace):
     """Build the conversation variable model for the API or Namespace."""
     return api_or_ns.model("ConversationVariable", conversation_variable_fields)
 
 
-def build_conversation_variable_infinite_scroll_pagination_model(api_or_ns: Api | Namespace):
+def build_conversation_variable_infinite_scroll_pagination_model(api_or_ns: Namespace):
     """Build the conversation variable infinite scroll pagination model for the API or Namespace."""
     # Build the nested variable model first
     conversation_variable_model = build_conversation_variable_model(api_or_ns)

api/fields/end_user_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 simple_end_user_fields = {
     "id": fields.String,
@@ -8,5 +8,5 @@ simple_end_user_fields = {
 }
 
 
-def build_simple_end_user_model(api_or_ns: Api | Namespace):
+def build_simple_end_user_model(api_or_ns: Namespace):
     return api_or_ns.model("SimpleEndUser", simple_end_user_fields)

api/fields/file_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from libs.helper import TimestampField
 
@@ -14,7 +14,7 @@ upload_config_fields = {
 }
 
 
-def build_upload_config_model(api_or_ns: Api | Namespace):
+def build_upload_config_model(api_or_ns: Namespace):
     """Build the upload config model for the API or Namespace.
 
     Args:
@@ -39,7 +39,7 @@ file_fields = {
 }
 
 
-def build_file_model(api_or_ns: Api | Namespace):
+def build_file_model(api_or_ns: Namespace):
     """Build the file model for the API or Namespace.
 
     Args:
@@ -57,7 +57,7 @@ remote_file_info_fields = {
 }
 
 
-def build_remote_file_info_model(api_or_ns: Api | Namespace):
+def build_remote_file_info_model(api_or_ns: Namespace):
     """Build the remote file info model for the API or Namespace.
 
     Args:
@@ -81,7 +81,7 @@ file_fields_with_signed_url = {
 }
 
 
-def build_file_with_signed_url_model(api_or_ns: Api | Namespace):
+def build_file_with_signed_url_model(api_or_ns: Namespace):
     """Build the file with signed URL model for the API or Namespace.
 
     Args:

api/fields/member_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from libs.helper import AvatarUrlField, TimestampField
 
@@ -9,7 +9,7 @@ simple_account_fields = {
 }
 
 
-def build_simple_account_model(api_or_ns: Api | Namespace):
+def build_simple_account_model(api_or_ns: Namespace):
     return api_or_ns.model("SimpleAccount", simple_account_fields)
 
 

api/fields/message_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from fields.conversation_fields import message_file_fields
 from libs.helper import TimestampField
@@ -10,7 +10,7 @@ feedback_fields = {
 }
 
 
-def build_feedback_model(api_or_ns: Api | Namespace):
+def build_feedback_model(api_or_ns: Namespace):
     """Build the feedback model for the API or Namespace."""
     return api_or_ns.model("Feedback", feedback_fields)
 
@@ -30,7 +30,7 @@ agent_thought_fields = {
 }
 
 
-def build_agent_thought_model(api_or_ns: Api | Namespace):
+def build_agent_thought_model(api_or_ns: Namespace):
     """Build the agent thought model for the API or Namespace."""
     return api_or_ns.model("AgentThought", agent_thought_fields)
 

api/fields/tag_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 dataset_tag_fields = {
     "id": fields.String,
@@ -8,5 +8,5 @@ dataset_tag_fields = {
 }
 
 
-def build_dataset_tag_fields(api_or_ns: Api | Namespace):
+def build_dataset_tag_fields(api_or_ns: Namespace):
     return api_or_ns.model("DataSetTag", dataset_tag_fields)

api/fields/workflow_app_log_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from fields.end_user_fields import build_simple_end_user_model, simple_end_user_fields
 from fields.member_fields import build_simple_account_model, simple_account_fields
@@ -17,7 +17,7 @@ workflow_app_log_partial_fields = {
 }
 
 
-def build_workflow_app_log_partial_model(api_or_ns: Api | Namespace):
+def build_workflow_app_log_partial_model(api_or_ns: Namespace):
     """Build the workflow app log partial model for the API or Namespace."""
     workflow_run_model = build_workflow_run_for_log_model(api_or_ns)
     simple_account_model = build_simple_account_model(api_or_ns)
@@ -43,7 +43,7 @@ workflow_app_log_pagination_fields = {
 }
 
 
-def build_workflow_app_log_pagination_model(api_or_ns: Api | Namespace):
+def build_workflow_app_log_pagination_model(api_or_ns: Namespace):
     """Build the workflow app log pagination model for the API or Namespace."""
     # Build the nested partial model first
     workflow_app_log_partial_model = build_workflow_app_log_partial_model(api_or_ns)

api/fields/workflow_run_fields.py

@@ -1,4 +1,4 @@
-from flask_restx import Api, Namespace, fields
+from flask_restx import Namespace, fields
 
 from fields.end_user_fields import simple_end_user_fields
 from fields.member_fields import simple_account_fields
@@ -19,7 +19,7 @@ workflow_run_for_log_fields = {
 }
 
 
-def build_workflow_run_for_log_model(api_or_ns: Api | Namespace):
+def build_workflow_run_for_log_model(api_or_ns: Namespace):
     return api_or_ns.model("WorkflowRunForLog", workflow_run_for_log_fields)
 
 

api/libs/archive_storage.py

@@ -0,0 +1,347 @@
+"""
+Archive Storage Client for S3-compatible storage.
+
+This module provides a dedicated storage client for archiving or exporting logs
+to S3-compatible object storage.
+"""
+
+import base64
+import datetime
+import gzip
+import hashlib
+import logging
+from collections.abc import Generator
+from typing import Any, cast
+
+import boto3
+import orjson
+from botocore.client import Config
+from botocore.exceptions import ClientError
+
+from configs import dify_config
+
+logger = logging.getLogger(__name__)
+
+
+class ArchiveStorageError(Exception):
+    """Base exception for archive storage operations."""
+
+    pass
+
+
+class ArchiveStorageNotConfiguredError(ArchiveStorageError):
+    """Raised when archive storage is not properly configured."""
+
+    pass
+
+
+class ArchiveStorage:
+    """
+    S3-compatible storage client for archiving or exporting.
+
+    This client provides methods for storing and retrieving archived data in JSONL+gzip format.
+    """
+
+    def __init__(self, bucket: str):
+        if not dify_config.ARCHIVE_STORAGE_ENABLED:
+            raise ArchiveStorageNotConfiguredError("Archive storage is not enabled")
+
+        if not bucket:
+            raise ArchiveStorageNotConfiguredError("Archive storage bucket is not configured")
+        if not all(
+            [
+                dify_config.ARCHIVE_STORAGE_ENDPOINT,
+                bucket,
+                dify_config.ARCHIVE_STORAGE_ACCESS_KEY,
+                dify_config.ARCHIVE_STORAGE_SECRET_KEY,
+            ]
+        ):
+            raise ArchiveStorageNotConfiguredError(
+                "Archive storage configuration is incomplete. "
+                "Required: ARCHIVE_STORAGE_ENDPOINT, ARCHIVE_STORAGE_ACCESS_KEY, "
+                "ARCHIVE_STORAGE_SECRET_KEY, and a bucket name"
+            )
+
+        self.bucket = bucket
+        self.client = boto3.client(
+            "s3",
+            endpoint_url=dify_config.ARCHIVE_STORAGE_ENDPOINT,
+            aws_access_key_id=dify_config.ARCHIVE_STORAGE_ACCESS_KEY,
+            aws_secret_access_key=dify_config.ARCHIVE_STORAGE_SECRET_KEY,
+            region_name=dify_config.ARCHIVE_STORAGE_REGION,
+            config=Config(s3={"addressing_style": "path"}),
+        )
+
+        # Verify bucket accessibility
+        try:
+            self.client.head_bucket(Bucket=self.bucket)
+        except ClientError as e:
+            error_code = e.response.get("Error", {}).get("Code")
+            if error_code == "404":
+                raise ArchiveStorageNotConfiguredError(f"Archive bucket '{self.bucket}' does not exist")
+            elif error_code == "403":
+                raise ArchiveStorageNotConfiguredError(f"Access denied to archive bucket '{self.bucket}'")
+            else:
+                raise ArchiveStorageError(f"Failed to access archive bucket: {e}")
+
+    def put_object(self, key: str, data: bytes) -> str:
+        """
+        Upload an object to the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+            data: Binary data to upload
+
+        Returns:
+            MD5 checksum of the uploaded data
+
+        Raises:
+            ArchiveStorageError: If upload fails
+        """
+        checksum = hashlib.md5(data).hexdigest()
+        try:
+            self.client.put_object(
+                Bucket=self.bucket,
+                Key=key,
+                Body=data,
+                ContentMD5=self._content_md5(data),
+            )
+            logger.debug("Uploaded object: %s (size=%d, checksum=%s)", key, len(data), checksum)
+            return checksum
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to upload object '{key}': {e}")
+
+    def get_object(self, key: str) -> bytes:
+        """
+        Download an object from the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Returns:
+            Binary data of the object
+
+        Raises:
+            ArchiveStorageError: If download fails
+            FileNotFoundError: If object does not exist
+        """
+        try:
+            response = self.client.get_object(Bucket=self.bucket, Key=key)
+            return response["Body"].read()
+        except ClientError as e:
+            error_code = e.response.get("Error", {}).get("Code")
+            if error_code == "NoSuchKey":
+                raise FileNotFoundError(f"Archive object not found: {key}")
+            raise ArchiveStorageError(f"Failed to download object '{key}': {e}")
+
+    def get_object_stream(self, key: str) -> Generator[bytes, None, None]:
+        """
+        Stream an object from the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Yields:
+            Chunks of binary data
+
+        Raises:
+            ArchiveStorageError: If download fails
+            FileNotFoundError: If object does not exist
+        """
+        try:
+            response = self.client.get_object(Bucket=self.bucket, Key=key)
+            yield from response["Body"].iter_chunks()
+        except ClientError as e:
+            error_code = e.response.get("Error", {}).get("Code")
+            if error_code == "NoSuchKey":
+                raise FileNotFoundError(f"Archive object not found: {key}")
+            raise ArchiveStorageError(f"Failed to stream object '{key}': {e}")
+
+    def object_exists(self, key: str) -> bool:
+        """
+        Check if an object exists in the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Returns:
+            True if object exists, False otherwise
+        """
+        try:
+            self.client.head_object(Bucket=self.bucket, Key=key)
+            return True
+        except ClientError:
+            return False
+
+    def delete_object(self, key: str) -> None:
+        """
+        Delete an object from the archive storage.
+
+        Args:
+            key: Object key (path) within the bucket
+
+        Raises:
+            ArchiveStorageError: If deletion fails
+        """
+        try:
+            self.client.delete_object(Bucket=self.bucket, Key=key)
+            logger.debug("Deleted object: %s", key)
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to delete object '{key}': {e}")
+
+    def generate_presigned_url(self, key: str, expires_in: int = 3600) -> str:
+        """
+        Generate a pre-signed URL for downloading an object.
+
+        Args:
+            key: Object key (path) within the bucket
+            expires_in: URL validity duration in seconds (default: 1 hour)
+
+        Returns:
+            Pre-signed URL string.
+
+        Raises:
+            ArchiveStorageError: If generation fails
+        """
+        try:
+            return self.client.generate_presigned_url(
+                ClientMethod="get_object",
+                Params={"Bucket": self.bucket, "Key": key},
+                ExpiresIn=expires_in,
+            )
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to generate pre-signed URL for '{key}': {e}")
+
+    def list_objects(self, prefix: str) -> list[str]:
+        """
+        List objects under a given prefix.
+
+        Args:
+            prefix: Object key prefix to filter by
+
+        Returns:
+            List of object keys matching the prefix
+        """
+        keys = []
+        paginator = self.client.get_paginator("list_objects_v2")
+
+        try:
+            for page in paginator.paginate(Bucket=self.bucket, Prefix=prefix):
+                for obj in page.get("Contents", []):
+                    keys.append(obj["Key"])
+        except ClientError as e:
+            raise ArchiveStorageError(f"Failed to list objects with prefix '{prefix}': {e}")
+
+        return keys
+
+    @staticmethod
+    def _content_md5(data: bytes) -> str:
+        """Calculate base64-encoded MD5 for Content-MD5 header."""
+        return base64.b64encode(hashlib.md5(data).digest()).decode()
+
+    @staticmethod
+    def serialize_to_jsonl_gz(records: list[dict[str, Any]]) -> bytes:
+        """
+        Serialize records to gzipped JSONL format.
+
+        Args:
+            records: List of dictionaries to serialize
+
+        Returns:
+            Gzipped JSONL bytes
+        """
+        lines = []
+        for record in records:
+            # Convert datetime objects to ISO format strings
+            serialized = ArchiveStorage._serialize_record(record)
+            lines.append(orjson.dumps(serialized))
+
+        jsonl_content = b"\n".join(lines)
+        if jsonl_content:
+            jsonl_content += b"\n"
+
+        return gzip.compress(jsonl_content)
+
+    @staticmethod
+    def deserialize_from_jsonl_gz(data: bytes) -> list[dict[str, Any]]:
+        """
+        Deserialize gzipped JSONL data to records.
+
+        Args:
+            data: Gzipped JSONL bytes
+
+        Returns:
+            List of dictionaries
+        """
+        jsonl_content = gzip.decompress(data)
+        records = []
+
+        for line in jsonl_content.splitlines():
+            if line:
+                records.append(orjson.loads(line))
+
+        return records
+
+    @staticmethod
+    def _serialize_record(record: dict[str, Any]) -> dict[str, Any]:
+        """Serialize a single record, converting special types."""
+
+        def _serialize(item: Any) -> Any:
+            if isinstance(item, datetime.datetime):
+                return item.isoformat()
+            if isinstance(item, dict):
+                return {key: _serialize(value) for key, value in item.items()}
+            if isinstance(item, list):
+                return [_serialize(value) for value in item]
+            return item
+
+        return cast(dict[str, Any], _serialize(record))
+
+    @staticmethod
+    def compute_checksum(data: bytes) -> str:
+        """Compute MD5 checksum of data."""
+        return hashlib.md5(data).hexdigest()
+
+
+# Singleton instance (lazy initialization)
+_archive_storage: ArchiveStorage | None = None
+_export_storage: ArchiveStorage | None = None
+
+
+def get_archive_storage() -> ArchiveStorage:
+    """
+    Get the archive storage singleton instance.
+
+    Returns:
+        ArchiveStorage instance
+
+    Raises:
+        ArchiveStorageNotConfiguredError: If archive storage is not configured
+    """
+    global _archive_storage
+    if _archive_storage is None:
+        archive_bucket = dify_config.ARCHIVE_STORAGE_ARCHIVE_BUCKET
+        if not archive_bucket:
+            raise ArchiveStorageNotConfiguredError(
+                "Archive storage bucket is not configured. Required: ARCHIVE_STORAGE_ARCHIVE_BUCKET"
+            )
+        _archive_storage = ArchiveStorage(bucket=archive_bucket)
+    return _archive_storage
+
+
+def get_export_storage() -> ArchiveStorage:
+    """
+    Get the export storage singleton instance.
+
+    Returns:
+        ArchiveStorage instance
+    """
+    global _export_storage
+    if _export_storage is None:
+        export_bucket = dify_config.ARCHIVE_STORAGE_EXPORT_BUCKET
+        if not export_bucket:
+            raise ArchiveStorageNotConfiguredError(
+                "Archive export bucket is not configured. Required: ARCHIVE_STORAGE_EXPORT_BUCKET"
+            )
+        _export_storage = ArchiveStorage(bucket=export_bucket)
+    return _export_storage

api/libs/external_api.py

@@ -1,5 +1,4 @@
 import re
-import sys
 from collections.abc import Mapping
 from typing import Any
 
@@ -109,11 +108,8 @@ def register_external_error_handlers(api: Api):
         data.setdefault("code", "unknown")
         data.setdefault("status", status_code)
 
-        # Log stack
-        exc_info: Any = sys.exc_info()
-        if exc_info[1] is None:
-            exc_info = (None, None, None)
-        current_app.log_exception(exc_info)
+        # Note: Exception logging is handled by Flask/Flask-RESTX framework automatically
+        # Explicit log_exception call removed to avoid duplicate log entries
 
         return data, status_code
 

api/migrations/versions/00bacef91f18_rename_api_provider_description.py

@@ -11,9 +11,6 @@ from alembic import op
 import models.types
 
 
-def _is_pg(conn):
-    return conn.dialect.name == "postgresql"
-
 # revision identifiers, used by Alembic.
 revision = '00bacef91f18'
 down_revision = '8ec536f3c800'
@@ -23,31 +20,17 @@ depends_on = None
 
 def upgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('description', sa.Text(), nullable=False))
-            batch_op.drop_column('description_str')
-    else:
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('description', models.types.LongText(), nullable=False))
-            batch_op.drop_column('description_str')
+    with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('description', models.types.LongText(), nullable=False))
+        batch_op.drop_column('description_str')
 
     # ### end Alembic commands ###
 
 
 def downgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('description_str', sa.TEXT(), autoincrement=False, nullable=False))
-            batch_op.drop_column('description')
-    else:
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('description_str', models.types.LongText(), autoincrement=False, nullable=False))
-            batch_op.drop_column('description')
+    with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('description_str', models.types.LongText(), autoincrement=False, nullable=False))
+        batch_op.drop_column('description')
 
     # ### end Alembic commands ###

api/migrations/versions/114eed84c228_remove_tool_id_from_model_invoke.py

@@ -7,14 +7,10 @@ Create Date: 2024-01-10 04:40:57.257824
 """
 import sqlalchemy as sa
 from alembic import op
-from sqlalchemy.dialects import postgresql
 
 import models.types
 
 
-def _is_pg(conn):
-    return conn.dialect.name == "postgresql"
-
 # revision identifiers, used by Alembic.
 revision = '114eed84c228'
 down_revision = 'c71211c8f604'
@@ -32,13 +28,7 @@ def upgrade():
 
 def downgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('tool_model_invokes', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('tool_id', postgresql.UUID(), autoincrement=False, nullable=False))
-    else:
-        with op.batch_alter_table('tool_model_invokes', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('tool_id', models.types.StringUUID(), autoincrement=False, nullable=False))
+    with op.batch_alter_table('tool_model_invokes', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('tool_id', models.types.StringUUID(), autoincrement=False, nullable=False))
 
     # ### end Alembic commands ###

api/migrations/versions/161cadc1af8d_add_dataset_permission_tenant_id.py

@@ -11,9 +11,6 @@ from alembic import op
 import models.types
 
 
-def _is_pg(conn):
-    return conn.dialect.name == "postgresql"
-
 # revision identifiers, used by Alembic.
 revision = '161cadc1af8d'
 down_revision = '7e6a8693e07a'
@@ -23,16 +20,9 @@ depends_on = None
 
 def upgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('dataset_permissions', schema=None) as batch_op:
-            # Step 1: Add column without NOT NULL constraint
-            op.add_column('dataset_permissions', sa.Column('tenant_id', sa.UUID(), nullable=False))
-    else:
-        with op.batch_alter_table('dataset_permissions', schema=None) as batch_op:
-            # Step 1: Add column without NOT NULL constraint
-            op.add_column('dataset_permissions', sa.Column('tenant_id', models.types.StringUUID(), nullable=False))
+    with op.batch_alter_table('dataset_permissions', schema=None) as batch_op:
+        # Step 1: Add column without NOT NULL constraint
+        op.add_column('dataset_permissions', sa.Column('tenant_id', models.types.StringUUID(), nullable=False))
 
     # ### end Alembic commands ###
 

api/migrations/versions/2024_09_24_0922-6af6a521a53e_update_retrieval_resource.py

@@ -9,11 +9,6 @@ from alembic import op
 import models.types
 
 
-def _is_pg(conn):
-    return conn.dialect.name == "postgresql"
-import sqlalchemy as sa
-from sqlalchemy.dialects import postgresql
-
 # revision identifiers, used by Alembic.
 revision = '6af6a521a53e'
 down_revision = 'd57ba9ebb251'
@@ -23,58 +18,30 @@ depends_on = None
 
 def upgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('dataset_retriever_resources', schema=None) as batch_op:
-            batch_op.alter_column('document_id',
-                   existing_type=sa.UUID(),
-                   nullable=True)
-            batch_op.alter_column('data_source_type',
-                   existing_type=sa.TEXT(),
-                   nullable=True)
-            batch_op.alter_column('segment_id',
-                   existing_type=sa.UUID(),
-                   nullable=True)
-    else:
-        with op.batch_alter_table('dataset_retriever_resources', schema=None) as batch_op:
-            batch_op.alter_column('document_id',
-                   existing_type=models.types.StringUUID(),
-                   nullable=True)
-            batch_op.alter_column('data_source_type',
-                   existing_type=models.types.LongText(),
-                   nullable=True)
-            batch_op.alter_column('segment_id',
-                   existing_type=models.types.StringUUID(),
-                   nullable=True)
+    with op.batch_alter_table('dataset_retriever_resources', schema=None) as batch_op:
+       batch_op.alter_column('document_id',
+              existing_type=models.types.StringUUID(),
+              nullable=True)
+       batch_op.alter_column('data_source_type',
+              existing_type=models.types.LongText(),
+              nullable=True)
+       batch_op.alter_column('segment_id',
+              existing_type=models.types.StringUUID(),
+              nullable=True)
     # ### end Alembic commands ###
 
 
 def downgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('dataset_retriever_resources', schema=None) as batch_op:
-            batch_op.alter_column('segment_id',
-                   existing_type=sa.UUID(),
-                   nullable=False)
-            batch_op.alter_column('data_source_type',
-                   existing_type=sa.TEXT(),
-                   nullable=False)
-            batch_op.alter_column('document_id',
-                   existing_type=sa.UUID(),
-                   nullable=False)
-    else:
-        with op.batch_alter_table('dataset_retriever_resources', schema=None) as batch_op:
-            batch_op.alter_column('segment_id',
-                   existing_type=models.types.StringUUID(),
-                   nullable=False)
-            batch_op.alter_column('data_source_type',
-                   existing_type=models.types.LongText(),
-                   nullable=False)
-            batch_op.alter_column('document_id',
-                   existing_type=models.types.StringUUID(),
-                   nullable=False)
+    with op.batch_alter_table('dataset_retriever_resources', schema=None) as batch_op:
+       batch_op.alter_column('segment_id',
+              existing_type=models.types.StringUUID(),
+              nullable=False)
+       batch_op.alter_column('data_source_type',
+              existing_type=models.types.LongText(),
+              nullable=False)
+       batch_op.alter_column('document_id',
+              existing_type=models.types.StringUUID(),
+              nullable=False)
 
     # ### end Alembic commands ###

api/migrations/versions/2024_11_01_0434-d3f6769a94a3_add_upload_files_source_url.py

@@ -8,7 +8,6 @@ Create Date: 2024-11-01 04:34:23.816198
 from alembic import op
 import models as models
 import sqlalchemy as sa
-from sqlalchemy.dialects import postgresql
 
 # revision identifiers, used by Alembic.
 revision = 'd3f6769a94a3'

api/migrations/versions/2024_11_01_0622-d07474999927_update_type_of_custom_disclaimer_to_text.py

@@ -28,85 +28,45 @@ def upgrade():
     op.execute("UPDATE sites SET custom_disclaimer = '' WHERE custom_disclaimer IS NULL")
     op.execute("UPDATE tool_api_providers SET custom_disclaimer = '' WHERE custom_disclaimer IS NULL")
 
-    if _is_pg(conn):
-        with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.VARCHAR(length=255),
-                   type_=sa.TEXT(),
-                   nullable=False)
+    with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
+        batch_op.alter_column('custom_disclaimer',
+                existing_type=sa.VARCHAR(length=255),
+                type_=models.types.LongText(),
+                nullable=False)
 
-        with op.batch_alter_table('sites', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.VARCHAR(length=255),
-                   type_=sa.TEXT(),
-                   nullable=False)
+    with op.batch_alter_table('sites', schema=None) as batch_op:
+        batch_op.alter_column('custom_disclaimer',
+                existing_type=sa.VARCHAR(length=255),
+                type_=models.types.LongText(),
+                nullable=False)
 
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.VARCHAR(length=255),
-                   type_=sa.TEXT(),
-                   nullable=False)
-    else:
-        with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.VARCHAR(length=255),
-                   type_=models.types.LongText(),
-                   nullable=False)
-
-        with op.batch_alter_table('sites', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.VARCHAR(length=255),
-                   type_=models.types.LongText(),
-                   nullable=False)
-
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.VARCHAR(length=255),
-                   type_=models.types.LongText(),
-                   nullable=False)
+    with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
+        batch_op.alter_column('custom_disclaimer',
+                existing_type=sa.VARCHAR(length=255),
+                type_=models.types.LongText(),
+                nullable=False)
 
     # ### end Alembic commands ###
 
 
 def downgrade():
     # ### commands auto generated by Alembic - please adjust! ###
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.TEXT(),
-                   type_=sa.VARCHAR(length=255),
-                   nullable=True)
+    with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
+        batch_op.alter_column('custom_disclaimer',
+                existing_type=models.types.LongText(),
+                type_=sa.VARCHAR(length=255),
+                nullable=True)
 
-        with op.batch_alter_table('sites', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.TEXT(),
-                   type_=sa.VARCHAR(length=255),
-                   nullable=True)
+    with op.batch_alter_table('sites', schema=None) as batch_op:
+        batch_op.alter_column('custom_disclaimer',
+                existing_type=models.types.LongText(),
+                type_=sa.VARCHAR(length=255),
+                nullable=True)
 
-        with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=sa.TEXT(),
-                   type_=sa.VARCHAR(length=255),
-                   nullable=True)
-    else:
-        with op.batch_alter_table('tool_api_providers', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=models.types.LongText(),
-                   type_=sa.VARCHAR(length=255),
-                   nullable=True)
-
-        with op.batch_alter_table('sites', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=models.types.LongText(),
-                   type_=sa.VARCHAR(length=255),
-                   nullable=True)
-
-        with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
-            batch_op.alter_column('custom_disclaimer',
-                   existing_type=models.types.LongText(),
-                   type_=sa.VARCHAR(length=255),
-                   nullable=True)
+    with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
+        batch_op.alter_column('custom_disclaimer',
+                existing_type=models.types.LongText(),
+                type_=sa.VARCHAR(length=255),
+                nullable=True)
 
     # ### end Alembic commands ###

api/migrations/versions/2024_11_01_0623-09a8d1878d9b_update_workflows_graph_features_and_.py

@@ -49,57 +49,33 @@ def upgrade():
     op.execute("UPDATE workflows SET updated_at = created_at WHERE updated_at IS NULL")
     op.execute("UPDATE workflows SET graph = '' WHERE graph IS NULL")
     op.execute("UPDATE workflows SET features = '' WHERE features IS NULL")
-    if _is_pg(conn):
-        with op.batch_alter_table('workflows', schema=None) as batch_op:
-            batch_op.alter_column('graph',
-                existing_type=sa.TEXT(),
-                nullable=False)
-            batch_op.alter_column('features',
-                existing_type=sa.TEXT(),
-                nullable=False)
-            batch_op.alter_column('updated_at',
-                existing_type=postgresql.TIMESTAMP(),
-                nullable=False)
-    else:
-        with op.batch_alter_table('workflows', schema=None) as batch_op:
-            batch_op.alter_column('graph',
-                existing_type=models.types.LongText(),
-                nullable=False)
-            batch_op.alter_column('features',
-                existing_type=models.types.LongText(),
-                nullable=False)
-            batch_op.alter_column('updated_at',
-                existing_type=sa.TIMESTAMP(),
-                nullable=False)
+
+    with op.batch_alter_table('workflows', schema=None) as batch_op:
+        batch_op.alter_column('graph',
+            existing_type=models.types.LongText(),
+            nullable=False)
+        batch_op.alter_column('features',
+            existing_type=models.types.LongText(),
+            nullable=False)
+        batch_op.alter_column('updated_at',
+            existing_type=sa.TIMESTAMP(),
+            nullable=False)
     # ### end Alembic commands ###
 
 
 def downgrade():
     # ### commands auto generated by Alembic - please adjust! ###
     conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('workflows', schema=None) as batch_op:
-            batch_op.alter_column('updated_at',
-                existing_type=postgresql.TIMESTAMP(),
-                nullable=True)
-            batch_op.alter_column('features',
-                existing_type=sa.TEXT(),
-                nullable=True)
-            batch_op.alter_column('graph',
-                existing_type=sa.TEXT(),
-                nullable=True)
-    else:
-        with op.batch_alter_table('workflows', schema=None) as batch_op:
-            batch_op.alter_column('updated_at',
-                existing_type=sa.TIMESTAMP(),
-                nullable=True)
-            batch_op.alter_column('features',
-                existing_type=models.types.LongText(),
-                nullable=True)
-            batch_op.alter_column('graph',
-                existing_type=models.types.LongText(),
-                nullable=True)
+    with op.batch_alter_table('workflows', schema=None) as batch_op:
+        batch_op.alter_column('updated_at',
+            existing_type=sa.TIMESTAMP(),
+            nullable=True)
+        batch_op.alter_column('features',
+            existing_type=models.types.LongText(),
+            nullable=True)
+        batch_op.alter_column('graph',
+            existing_type=models.types.LongText(),
+            nullable=True)
 
     if _is_pg(conn):
         with op.batch_alter_table('messages', schema=None) as batch_op:

api/migrations/versions/2025_08_13_1605-0e154742a5fa_add_provider_model_multi_credential.py

@@ -86,57 +86,30 @@ def upgrade():
 
 def migrate_existing_provider_models_data():
     """migrate provider_models table data to provider_model_credentials"""
-    conn = op.get_bind()
-    # Define table structure for data manipulation
-    if _is_pg(conn):
-        provider_models_table = table('provider_models',
-            column('id', models.types.StringUUID()),
-            column('tenant_id', models.types.StringUUID()),
-            column('provider_name', sa.String()),
-            column('model_name', sa.String()),
-            column('model_type', sa.String()),
-            column('encrypted_config', sa.Text()),
-            column('created_at', sa.DateTime()),
-            column('updated_at', sa.DateTime()),
-            column('credential_id', models.types.StringUUID()),
-        )
-    else:
-        provider_models_table = table('provider_models',
-            column('id', models.types.StringUUID()),
-            column('tenant_id', models.types.StringUUID()),
-            column('provider_name', sa.String()),
-            column('model_name', sa.String()),
-            column('model_type', sa.String()),
-            column('encrypted_config', models.types.LongText()),
-            column('created_at', sa.DateTime()),
-            column('updated_at', sa.DateTime()),
-            column('credential_id', models.types.StringUUID()),
-        )
+    # Define table structure for data manipulatio
+    provider_models_table = table('provider_models',
+        column('id', models.types.StringUUID()),
+        column('tenant_id', models.types.StringUUID()),
+        column('provider_name', sa.String()),
+        column('model_name', sa.String()),
+        column('model_type', sa.String()),
+        column('encrypted_config', models.types.LongText()),
+        column('created_at', sa.DateTime()),
+        column('updated_at', sa.DateTime()),
+        column('credential_id', models.types.StringUUID()),
+    )
 
-    if _is_pg(conn):
-        provider_model_credentials_table = table('provider_model_credentials',
-            column('id', models.types.StringUUID()),
-            column('tenant_id', models.types.StringUUID()),
-            column('provider_name', sa.String()),
-            column('model_name', sa.String()),
-            column('model_type', sa.String()),
-            column('credential_name', sa.String()),
-            column('encrypted_config', sa.Text()),
-            column('created_at', sa.DateTime()),
-            column('updated_at', sa.DateTime())
-        )
-    else:
-        provider_model_credentials_table = table('provider_model_credentials',
-            column('id', models.types.StringUUID()),
-            column('tenant_id', models.types.StringUUID()),
-            column('provider_name', sa.String()),
-            column('model_name', sa.String()),
-            column('model_type', sa.String()),
-            column('credential_name', sa.String()),
-            column('encrypted_config', models.types.LongText()),
-            column('created_at', sa.DateTime()),
-            column('updated_at', sa.DateTime())
-        )
+    provider_model_credentials_table = table('provider_model_credentials',
+        column('id', models.types.StringUUID()),
+        column('tenant_id', models.types.StringUUID()),
+        column('provider_name', sa.String()),
+        column('model_name', sa.String()),
+        column('model_type', sa.String()),
+        column('credential_name', sa.String()),
+        column('encrypted_config', models.types.LongText()),
+        column('created_at', sa.DateTime()),
+        column('updated_at', sa.DateTime())
+    )
 
 
     # Get database connection
@@ -183,14 +156,8 @@ def migrate_existing_provider_models_data():
 
 def downgrade():
     # Re-add encrypted_config column to provider_models table
-    conn = op.get_bind()
-    
-    if _is_pg(conn):
-        with op.batch_alter_table('provider_models', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('encrypted_config', sa.Text(), nullable=True))
-    else:
-        with op.batch_alter_table('provider_models', schema=None) as batch_op:
-            batch_op.add_column(sa.Column('encrypted_config', models.types.LongText(), nullable=True))
+    with op.batch_alter_table('provider_models', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('encrypted_config', models.types.LongText(), nullable=True))
 
     if not context.is_offline_mode():
         # Migrate data back from provider_model_credentials to provider_models

api/migrations/versions/2025_08_20_1747-8d289573e1da_add_oauth_provider_apps.py

@@ -8,7 +8,6 @@ Create Date: 2025-08-20 17:47:17.015695
 from alembic import op
 import models as models
 import sqlalchemy as sa
-from libs.uuid_utils import uuidv7
 
 
 def _is_pg(conn):