7 Commits

Author	SHA1	Message	Date
Lars	a1723db387	feat: Workflow Engine Part 3 - Inline Prompts (v0.9q) ... Ermöglicht Analysis Nodes zwischen zwei Prompt-Modi zu wählen: - Reference Mode: Basis-Prompt aus DB referenzieren (bestehend) - Inline Mode: Template direkt im Node editieren (NEU) Frontend: - InlineTemplateEditor Component (~80 Zeilen) - Radio Buttons in WorkflowEditorPage für Mode-Auswahl - Placeholder Picker für beide Modi (End Node + Inline Template) - Cursor-Position Tracking mit textareaRef - Conditional Rendering basierend auf promptSource - Validation: Entweder prompt_slug ODER inline_template Backend: - load_prompt_template() akzeptiert ganzen WorkflowNode (statt nur slug) - Unterstützt inline_template (Mode 1) und prompt_slug (Mode 2) - WorkflowNode.inline_template Feld hinzugefügt - Validation: HTTPException wenn weder slug noch template Serialization: - inline_template in graph_data speichern/laden - Backward-compatible mit bestehenden Workflows Version: 0.9q Module: workflow 0.7.0 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-11 08:45:00 +02:00
Lars	856a82ec1d	fix: Frontend-Backend field name mismatch for workflow questions ... Root Cause: - Frontend serialized as "questions" - Backend expected "question_augmentations" - Analysis Nodes WITH questions configured sent empty array to backend - Questions were never added to LLM prompt Frontend workflowSerializer.js: - Serialization: questions → question_augmentations (Backend field name) - Deserialization: question_augmentations → questions (Frontend data object) - Backward compatible: Falls back to "questions" for old workflows Backend workflow_executor.py: - Removed incorrect load_prompt_questions() function (was a misunderstanding) - Back to original logic: Only use node.question_augmentations - Simplified normalization logging Impact: - Analysis Node questions are now correctly sent to backend - Questions augment the base prompt as intended - LLM receives structured questions - Decision signals are generated and accessible as placeholders Example: - Node configures question with id="q21" - Signal becomes accessible as {{ node_2.signal_q21 }} - Can be used in Logic Nodes and End Node templates Issue: Workflow questions not sent to LLM (field name mismatch) Version: 0.9p (workflow module) Part 3: End Node Template Engine - Critical Fix Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-09 18:28:54 +02:00
Lars	228010a6d3	feat: Part 3 - End Node Template Editor ... **Neue Features:** - End Node Output Mode: AUTO vs. TEMPLATE - Jinja2 Template Editor mit Syntax-Beispiel - Placeholder Picker Modal (dynamische Node-Liste) - Template Serialisierung/Deserialisierung **Komponenten (NEU):** 1. EndNodeConfig.jsx (~150 Zeilen) - Output Mode Toggle (AUTO/TEMPLATE) - Template Textarea (monospace, 12 Zeilen) - Placeholder-Button (öffnet Picker) - Help-Text mit Beispiel-Syntax - Auto-Insert Default Template beim Wechsel zu TEMPLATE 2. PlaceholderPicker.jsx (~260 Zeilen) - Modal mit Suchfunktion - Dynamische Placeholder-Liste aus Workflow-Nodes - Kategorien: Global, Node Outputs, Signals - Click-to-Insert (schließt Modal automatisch) - Icons pro Node-Typ (🚀🤖⚡🔀🏁) **Integration:** - WorkflowEditorPage.jsx - EndNodeConfig im Config Panel (wenn type='end') - PlaceholderPicker State + Modal - handlePlaceholderSelect (fügt in Template ein) **Serialisierung:** - workflowSerializer.js - Serialize: output_mode + template für End Nodes - Deserialize: output_mode + template laden - Fallback: auto Mode wenn nicht gesetzt **Backend Status:** - ✅ Backend bereits fertig (execute_end_node() in workflow_executor.py) - ✅ Beide Modi (AUTO/TEMPLATE) funktionieren - ✅ Jinja2 Template Rendering implementiert **Part 3 Status:** Frontend Complete - ✅ End Node Config UI - ✅ Template Editor - ✅ Placeholder Picker - ⏸️ Testing ausstehend **Nächster Schritt:** Browser-Test auf dev.mitai.jinkendo.de Version: v0.9p Date: 2026-04-09 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-09 15:52:19 +02:00
Lars	2994df54ad	fix: Complete prompt_id → prompt_slug migration ... **Remaining Issues Found:** 1. Validierung prüfte noch prompt_id → Speichern blockiert 2. AnalysisNode zeigte noch prompt_id → "Kein Prompt" obwohl gesetzt 3. Deserializer hatte keinen Fallback für alte Workflows **Fixes:** - workflowValidation.js: Line 91 - if (!node.data.prompt_slug) statt prompt_id - AnalysisNode.jsx: Line 15 - Verwendet jetzt data.prompt_slug - Fallback: "Prompt: ${slug}" statt "Prompt #${id}" - Kommentar aktualisiert (data.prompt_slug statt prompt_id) - workflowSerializer.js: Line 81 - Fallback: node.prompt_slug || node.prompt_id || null - Backwards-compatible für alte Workflows **Testing:** 1. Neuen Workflow erstellen 2. Analysis Node hinzufügen 3. Prompt auswählen 4. Node sollte Prompt-Namen zeigen 5. Beim Öffnen: Dropdown sollte Wert anzeigen 6. Speichern sollte funktionieren (keine Validierungsfehler) 7. Execute sollte funktionieren (kein "Prompt not found") Part 2 Complete Bugfix - Workflow Execution Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-09 15:39:11 +02:00
Lars	af7c5ca55a	fix: Workflow execute - prompt_id → prompt_slug ... **Problem:** Execute-Button funktionierte, aber Ausführung scheiterte mit "Prompt not found: none" im Analysis Node. **Root Cause:** Mismatch zwischen Frontend (speicherte prompt_id/UUID) und Backend (erwartete prompt_slug/String). Backend WorkflowNode Model (workflow_models.py:193): prompt_slug: Optional[str] = Field(...) **Änderungen:** - WorkflowEditorPage.jsx: Dropdown auf slug umgestellt - value: prompt_slug statt prompt_id - onChange: selectedPrompt.slug statt .id - handleNodeUpdate: prompt_slug speichern - workflowSerializer.js: Serialization/Deserialization - Serialize: prompt_slug statt prompt_id - Deserialize: node.prompt_slug lesen **Testing:** - Workflow mit Analysis Node neu erstellen - Execute-Button sollte jetzt funktionieren Part 2 Bugfix - Workflow Execution Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-09 13:59:54 +02:00
Lars	cab5758b0d	fix: Save prompt_name in graph_data for readable node display ... Bug: - Analysis nodes showed "Prompt #7edc6d6b-6cd5..." in canvas - After re-selecting prompt, showed readable name "Pipeline: Ernährungs-Analyse" - After loading workflow, showed UUID again Root Cause: - Serializer saved only prompt_id, not prompt_name - Deserializer expected prompt_name but got null - AnalysisNode fallback logic: data.prompt_name || `Prompt #{prompt_id}` - Result: Showed UUID as fallback Fix: - workflowSerializer.js line 25: Added prompt_name to serialization - Now saves both prompt_id AND prompt_name in graph_data - On load: prompt_name is restored → AnalysisNode shows readable name Testing: - Create workflow with analysis node + prompt selection - Save → Canvas should show "Pipeline: Körper-Analyse" (not UUID) - Reload → Canvas should still show readable name (not UUID)	2026-04-04 22:50:40 +02:00
Lars	dc59596f01	feat: Phase 5 - Visual Workflow Editor (Option B) ... Backend (Mini-Backend 1-2h): - Migration 016: ai_prompts.graph_data JSONB column - workflow_executor: graph_data parameter support (backward-compatible) - prompt_executor: execute_workflow_prompt uses graph_data Frontend (Main effort 25-35h): - WorkflowCanvas: React Flow wrapper component - 5 Custom Nodes: Start, End, Analysis, Logic, Join - 4 Config Panels: QuestionAugmentation, LogicExpression, Fallback, Join - workflowValidation: Structural + logical validation - workflowSerializer: Canvas ↔ JSONB conversion - WorkflowEditorPage: Main orchestration (420 LOC) - Route: /workflow-editor/:id - CSS: workflowEditor.css (300 LOC) Architecture: - Option B: ai_prompts.type='workflow' (not separate table) - panels/ subdirectory for clean separation - WorkflowCanvas reusable component - User GUI identical (Workflows = Prompts) - Backward-compatible (type='pipeline' unchanged) Version: v0.9m → v0.9n (Phase 5 complete) Module: workflow 0.5.0 → 0.6.0 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-04-04 17:56:00 +02:00