Self-improving AI Employees for everyone
Python ai-agentsai-employeeai-employeesassistantautonomous-agentsco-employeescompanygrok-botinstinctloop-agentsmuseown-your-datapersonalself-hosted
fix: Persist skillsConfig on user skill delete to prevent reappearance
trohitg02176488e4f4
diff --git a/.github/workflows/release.yml b/.github/workflows/release.ymlindex 3aac9d24..1685c57c 100644--- a/.github/workflows/release.yml+++ b/.github/workflows/release.yml@@ -30,6 +30,9 @@ jobs:- name: Install dependenciesrun: npm install --ignore-scripts+ - name: Sync versions from git tag+ run: node scripts/sync-version.js+- name: Verify package contentsrun: npm pack --dry-run@@ -58,6 +61,9 @@ jobs:- name: Install dependenciesrun: npm install --ignore-scripts+ - name: Sync versions from git tag+ run: node scripts/sync-version.js+- name: Configure scoped packagerun: |node -e "diff --git a/CLAUDE.md b/CLAUDE.mdindex dfbd2b95..9f3c303e 100644--- a/CLAUDE.md+++ b/CLAUDE.md@@ -87,7 +87,7 @@ server/core/server/models/├── cache.py # CacheEntry SQLModel for SQLite cache├── auth.py # User model with bcrypt-└── database.py # ConversationMessage, NodeParameter, ToolSchema, ChatMessage, TokenUsageMetric, CompactionEvent, SessionTokenState tables+└── database.py # ConversationMessage, NodeParameter, ToolSchema, ChatMessage, TokenUsageMetric, CompactionEvent, SessionTokenState, UserSettings, ProviderDefaults tables```### Polyglot Server Integration (Optional)@@ -290,15 +290,15 @@ class CacheEntry(SQLModel, table=True):## Codebase Summary- **Hybrid architecture**: Node.js + Python + React TypeScript-- **75 implemented workflow nodes** with clean service separation (6 AI models + 3 AI agents/memory + 10 specialized agents + 11 skills + 3 dedicated tools + 6 dual-purpose tools + 16 Android + 3 WhatsApp + 2 Social + 3 Location + 2 Code + 5 Utility + 6 Document + 2 Chat + 2 Scheduler + 1 Workflow)-- **WebSocket-First Architecture**: WebSocket as primary frontend-backend communication (84 message handlers)+- **77 implemented workflow nodes** with clean service separation (6 AI models + 3 AI agents/memory + 12 specialized agents + 11 skills + 3 dedicated tools + 6 dual-purpose tools + 16 Android + 3 WhatsApp + 2 Social + 3 Location + 2 Code + 5 Utility + 6 Document + 2 Chat + 2 Scheduler + 1 Workflow)+- **WebSocket-First Architecture**: WebSocket as primary frontend-backend communication (86 message handlers)- **Recent optimizations**: REST APIs replaced with WebSocket, AI endpoints migrated to Python, Android automation integrated## Architecture RefactoringThe project was completely refactored from schema-based node definitions to explicit INodeProperties interface system inspired by n8n architecture. Key changes:- **Pure INodeProperties System**: Removed all backward compatibility layers- **75 Implemented Node Components**: AI models, agents, skills, location services, Android automation, WhatsApp, social, code execution, HTTP/Webhook utilities, document processing, chat, and schedulers-- **WebSocket-First Communication**: 84 WebSocket message handlers replace most REST API calls+- **WebSocket-First Communication**: 86 WebSocket message handlers replace most REST API calls- **Resource-Operation Pattern**: Organized by functional categories (AI, Location, Android, WhatsApp)- **TypeScript-First**: Full type safety with proper interface alignment- **Code Cleanup**: Removed dead code, unused files, and legacy methods@@ -437,7 +437,7 @@ React Flow edges use dynamic Dracula colors via `getEdgeStyles(theme.dracula)`:### WebSocket-First ArchitectureThe project uses WebSocket as the primary communication method between frontend and backend, replacing most REST API calls:- `src/contexts/WebSocketContext.tsx` - Central WebSocket context with request/response pattern-- `server/routers/websocket.py` - WebSocket endpoint with 84 message handlers+- `server/routers/websocket.py` - WebSocket endpoint with 86 message handlers- `server/services/status_broadcaster.py` - Connection management and broadcasting## Implemented Node Types@@ -454,21 +454,12 @@ The following 75 nodes are currently implemented and functional:### AI Agents & Memory (3 nodes)- **aiAgent**: Advanced AI agent with tool calling, memory input handle, and iterative reasoning. Uses LangGraph for structured execution. Parameters: Provider, Model, Prompt, System Message, Options.- **chatAgent**: Conversational AI agent with memory and skill support for multi-turn chat interactions. Parameters: Provider, Model, Prompt (supports `{{chatTrigger.message}}` template or auto-fallback from connected input), System Message. Behavior extended by connected skills.-- **simpleMemory**: Markdown-based conversation memory with editable UI, window-based trimming, and optional vector DB for long-term semantic retrieval+### AI Agent Tool Nodes (5 dedicated + 6 dual-purpose)+Tool nodes connect to AI Agent's `input-tools` handle to provide capabilities the agent can invoke during reasoning. Both `masterSkill` and `simpleMemory` are in the AI Tools category.-### Zeenie Skill Nodes (11 nodes)-Skill nodes connect to Zeenie's `input-skill` handle to provide capabilities defined in SKILL.md files following the [Agent Skills specification](https://agentskills.io/specification):-- **masterSkill**: Master Skill (icon: target) - Aggregates multiple skills with enable/disable toggles. Split-panel UI: left panel shows skill list with checkboxes, right panel shows selected skill's markdown editor. Enabled skills are expanded into individual skill entries when the AI Agent executes.-- **assistantPersonality**: Assistant Skill (icon: sparkles) - default assistant personality with helpful, harmless, honest responses-- **whatsappSkill**: Send and receive WhatsApp messages via Zeenie-- **memorySkill**: Long-term memory management across conversations-- **mapsSkill**: Location services - geocoding, nearby places, directions via Google Maps-- **httpSkill**: Make HTTP requests to external APIs-- **schedulerSkill**: Task scheduling with timers and cron expressions-- **androidSkill**: Android device control - battery, wifi, bluetooth, apps, location, camera-- **codeSkill**: Execute Python or JavaScript code for calculations and data processing-- **webSearchSkill**: Search the web for current information, news, facts, and real-time data-- **customSkill**: User-created custom skill with configurable capabilities (select from database)+#### Dedicated Tool Nodes (passive, tool-only)+- **masterSkill**: Master Skill (icon: target) - Aggregates multiple skills with enable/disable toggles. Split-panel UI: left panel shows skill list with checkboxes, right panel shows selected skill's markdown editor. Supports both built-in skills (from `server/skills/` folders) and user-created skills (stored in database). User skills can be created/edited/deleted inline via the "+" button.+- **simpleMemory**: Markdown-based conversation memory with editable UI, window-based trimming, and optional vector DB for long-term semantic retrieval#### Skill Node ArchitectureSkills are organized in subfolders under `server/skills/`. Each top-level folder appears as an option in the Master Skill node's folder dropdown. See **[Skill Creation Guide](./server/skills/GUIDE.md)** for full documentation.@@ -607,20 +598,19 @@ if skill_type == 'masterSkill':**Key Files:**| File | Description ||------|-------------|-| `client/src/components/parameterPanel/MasterSkillEditor.tsx` | Split-panel skill aggregator UI with folder dropdown |-| `client/src/nodeDefinitions/skillNodes.ts` | masterSkill node definition |-| `server/services/handlers/ai.py` | Expands masterSkill into individual skills |-| `server/skills/GUIDE.md` | Skill creation guide |--### AI Agent Tool Nodes (3 dedicated + 6 dual-purpose)-Tool nodes connect to AI Agent's `input-tools` handle to provide capabilities the agent can invoke during reasoning.--#### Dedicated Tool Nodes (passive, tool-only)+| `client/src/components/parameterPanel/MasterSkillEditor.tsx` | Split-panel skill aggregator UI with inline user skill CRUD |+| `client/src/nodeDefinitions/skillNodes.ts` | masterSkill node definition (group: tool) |+| `server/routers/websocket.py` | User skill CRUD: `get_user_skills`, `create_user_skill`, `update_user_skill`, `delete_user_skill` |+| `server/core/database.py` | UserSkill model and database CRUD methods |+| `server/services/handlers/ai.py` | Expands masterSkill into individual skills at execution |+| `server/skills/GUIDE.md` | Skill creation guide for built-in skills |++#### Other Dedicated Tool Nodes- **calculatorTool**: Mathematical operations (add, subtract, multiply, divide, power, sqrt, mod, abs)- **currentTimeTool**: Get current date/time with timezone support- **webSearchTool**: Web search via DuckDuckGo (free, uses `ddgs` library) or Serper API with configurable max results-### Specialized AI Agents (10 nodes)+### Specialized AI Agents (12 nodes)Specialized agents are AI Agents pre-configured for specific domains. They inherit full AI Agent functionality (provider, model, prompt, system message, thinking/reasoning) while being tailored for specific capabilities. All specialized agents route to `handle_chat_agent` in the backend and support the same input handles. Node colors use centralized dracula theme constants imported from `client/src/styles/theme.ts`.**Input Handles:**@@ -641,6 +631,8 @@ Specialized agents are AI Agents pre-configured for specific domains. They inher- **productivity_agent**: Productivity Agent - AI agent for productivity workflows. Connect scheduling, task, and utility nodes as tools.- **payments_agent**: Payments Agent - AI agent for payment processing. Connect payment, invoice, and financial tool nodes.- **consumer_agent**: Consumer Agent - AI agent for consumer interactions. Connect customer support, product, and order management tools.+- **autonomous_agent**: Autonomous Agent - AI agent for autonomous operations using Code Mode patterns. Uses agentic loops, progressive discovery, error recovery, and multi-tool orchestration for 81-98% token savings. Connect autonomous skills via Master Skill.+- **orchestrator_agent**: Orchestrator Agent - AI agent for coordinating multiple agents in complex multi-agent workflows. Delegates tasks to specialized agents and synthesizes their results.**Backend Routing:**Specialized agents are detected by `SPECIALIZED_AGENT_TYPES` and routed to `handle_chat_agent`:@@ -648,7 +640,8 @@ Specialized agents are detected by `SPECIALIZED_AGENT_TYPES` and routed to `hand# In node_executor.py - all specialized agents route to handle_chat_agentSPECIALIZED_AGENT_TYPES = {'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent',- 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent'+ 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent',+ 'autonomous_agent', 'orchestrator_agent'}```@@ -973,7 +966,7 @@ See **[Scripts Reference](./docs-internal/SCRIPTS.md)** for full documentation.## Current Status✅ **INodeProperties System**: Fully implemented with 75 functional node components-✅ **WebSocket-First Architecture**: 84 message handlers replacing REST APIs+✅ **WebSocket-First Architecture**: 86 message handlers replacing REST APIs✅ **Code Editor**: Python executor with syntax-highlighted editor (react-simple-code-editor + prismjs) and console output✅ **Component Palette**: Emoji icons with distinct dracula-themed category colors, localStorage persistence for collapsed sections✅ **Android Integration**: 16 Android service nodes with ADB automation and remote WebSocket support@@ -1281,11 +1274,13 @@ deploy_workflow() -> Sets up triggers, returns immediately- **Execution Engine**: Routes AI nodes to Python Flask backend with auto-injection of API keys#### Supported AI Providers & Models-- **OpenAI**: GPT-4o, GPT-4 Turbo, GPT-3.5 Turbo with response format options. O-series (o1, o3, o4) with reasoning effort.-- **Anthropic**: Claude 3.5 Sonnet/Haiku, Claude 3 Opus with extended thinking support (thinkingBudget parameter)-- **Google**: Gemini 2.5 Pro/Flash, Gemini 2.0 Flash Thinking with thinking_budget parameter and safety settings-- **Groq**: Ultra-fast Llama, Mixtral, Gemma models. Qwen3/QwQ with reasoning_format parameter.-- **Cerebras**: Ultra-fast Llama and Qwen models on custom AI hardware with thinking support+All 6 providers are available across aiAgent, chatAgent (Zeenie), and all specialized agents:+- **OpenAI**: GPT-5, GPT-4o, o-series (o1, o3, o4) with reasoning effort. Max output: 128K tokens.+- **Anthropic**: Claude Opus 4.6, Sonnet 4.5 with extended thinking. Max output: 128K tokens.+- **Google**: Gemini 3.0, 2.5 Pro/Flash with thinking_budget. Max output: 65K tokens.+- **Groq**: Ultra-fast Llama, Qwen3/QwQ with reasoning_format. Max output: 8K tokens.+- **OpenRouter**: Unified API for 200+ models from multiple providers.+- **Cerebras**: Ultra-fast Llama and Qwen models on custom AI hardware.#### Key Features- Visual configuration with status indicators and parameter buttons@@ -1294,6 +1289,39 @@ deploy_workflow() -> Sets up triggers, returns immediately- Secure API key validation with automatic model discovery and 30-day expiration- Execution routing to Python Flask backend for AI model processing- **Proxy-based authentication** for routing through local servers (Ollama pattern)+- **Provider default parameters** configurable in Credentials Modal (temperature, max_tokens, thinking settings)++#### Provider Default Parameters+Users can configure default parameter values per LLM provider in the Credentials Modal. These defaults are applied to new AI nodes using that provider.++**Configurable Parameters:**+- `temperature` (0-2): Controls randomness in responses+- `max_tokens` (1-128000): Maximum response length+- `thinking_enabled`: Enable extended thinking for supported models+- `thinking_budget` (1024-16000): Token budget for thinking+- `reasoning_effort` (low/medium/high): For OpenAI o-series, Groq+- `reasoning_format` (parsed/hidden): For Groq Qwen models++**Database Model** (`server/models/database.py`):+```python+class ProviderDefaults(SQLModel, table=True):+ provider: str # openai, anthropic, gemini, groq, openrouter, cerebras+ temperature: float+ max_tokens: int+ thinking_enabled: bool+ thinking_budget: int+ reasoning_effort: str # low, medium, high+ reasoning_format: str # parsed, hidden+```++**Key Files:**+| File | Description |+|------|-------------|+| `server/models/database.py` | `ProviderDefaults` SQLModel |+| `server/core/database.py` | `get_provider_defaults()`, `save_provider_defaults()` CRUD |+| `server/routers/websocket.py` | `get_provider_defaults`, `save_provider_defaults` handlers |+| `client/src/hooks/useApiKeys.ts` | `getProviderDefaults()`, `saveProviderDefaults()` methods |+| `client/src/components/CredentialsModal.tsx` | Default Parameters UI section |#### Proxy-Based Authentication (Ollama Pattern)AI providers support optional proxy-based authentication, allowing requests to route through a local proxy server that handles authentication. This follows the [Ollama Claude Code integration](https://docs.ollama.com/integrations/claude-code) pattern.@@ -1366,10 +1394,10 @@ Extended thinking and reasoning capabilities for supported AI models. When enabl| Provider | Models | Parameter | Notes ||----------|--------|-----------|-------|-| **Claude** | claude-3-5-sonnet, claude-3-opus | `thinkingBudget` (1024-16000 tokens) | Requires `max_tokens > budget_tokens`. Temperature auto-set to 1. |-| **Gemini** | gemini-2.5-flash, gemini-2.5-pro, gemini-2.0-flash-thinking | `thinkingBudget` (token count) | Uses `thinking_budget` API parameter |+| **Claude** | opus-4.6, sonnet-4.5 | `thinkingBudget` (1024-16000 tokens) | Requires `max_tokens > budget_tokens`. Temperature auto-set to 1. |+| **Gemini** | gemini-3.0, gemini-2.5-pro/flash | `thinkingBudget` (token count) | Uses `thinking_budget` API parameter || **Groq** | qwen3-32b, qwq-32b | `reasoningFormat` ('parsed' or 'hidden') | 'parsed' returns reasoning, 'hidden' returns only final answer |-| **OpenAI** | o1, o3, o4 series | `reasoningEffort` (minimal/low/medium/high) | Reasoning summaries require organization verification at platform.openai.com |+| **OpenAI** | o1, o3, o4, GPT-5 series | `reasoningEffort` (minimal/low/medium/high/xhigh) | GPT-5.2 supports xhigh reasoning |#### Frontend Parameters (`client/src/factories/baseChatModelFactory.ts`)```typescript@@ -1521,8 +1549,8 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-task', label: 'Task', position: '85%' },],},- // ... 10 specialized agents (android_agent, coding_agent, web_agent, task_agent,- // social_agent, travel_agent, tool_agent, productivity_agent, payments_agent, consumer_agent)+ // ... 11 specialized agents (android_agent, coding_agent, web_agent, task_agent,+ // social_agent, travel_agent, tool_agent, productivity_agent, payments_agent, consumer_agent, autonomous_agent)// each with themeColorKey, standard Skill/Tool bottom handles, Memory/Task left handles};```@@ -1677,6 +1705,8 @@ The system supports specialized agent variants that inherit from the base AI Age| Productivity | `productivity_agent` | clock | cyan || Payments | `payments_agent` | credit card | green || Consumer | `consumer_agent` | cart | purple |+| Autonomous | `autonomous_agent` | target | purple |+| Orchestrator | `orchestrator_agent` | conductor | cyan |All specialized agents share the same handle configuration:- **Left**: `input-main` (Input, 30%), `input-memory` (Memory, 55%), `input-task` (Task, 85%)@@ -3875,7 +3905,7 @@ def has_real_android_devices(self) -> bool:Updated `client_left` and presence handlers use `has_real_android_devices()` instead of checking total device count.-### WebSocket Message Types (84 Handlers)+### WebSocket Message Types (86 Handlers)#### Request/Response Messages (Client -> Server -> Client)| Category | Message Types |@@ -3900,6 +3930,7 @@ Updated `client_left` and presence handlers use `has_real_android_devices()` ins| **Built-in Skills** | `get_skill_content`, `save_skill_content`, `scan_skill_folder`, `list_skill_folders` || **Memory/Skill Reset** | `clear_memory`, `reset_skill` || **User Settings** | `get_user_settings`, `save_user_settings` |+| **Provider Defaults** | `get_provider_defaults`, `save_provider_defaults` |#### Broadcast Messages (Server -> All Clients)| Message Type | Description |@@ -4056,7 +4087,7 @@ This function:- **Performance**: Fast HMR updates and clean TypeScript compilation- **AI Architecture**: 5-layer system with factory pattern and secure credential management- **Android Architecture**: Factory-based node creation with ADB integration for device automation-- **WebSocket-First Architecture**: 84 message handlers replace REST APIs for parameters, execution, API keys, Android, WhatsApp, and skill operations+- **WebSocket-First Architecture**: 86 message handlers replace REST APIs for parameters, execution, API keys, Android, WhatsApp, and skill operations- **WebSocket Hooks**: Dedicated React hooks (useWhatsApp, useExecution, useApiKeys, useAndroidOperations, useParameterPanel) for clean component integration- **WebSocket Support**: Persistent remote Android device connections via WebSocket proxy with background tasks- Connection stays alive across multiple API requests until switched to local ADBdiff --git a/README.md b/README.mdindex 2cf27d6e..ff5e5509 100644--- a/README.md+++ b/README.md@@ -57,11 +57,11 @@ npm run docker:up| Provider | Models | Features ||----------|--------|----------|-| **OpenAI** | GPT-4o, GPT-4 Turbo, o1, o3, o4-mini | JSON mode, reasoning effort |-| **Anthropic** | Claude 3.5 Sonnet, Claude 3 Opus, Claude 3 Haiku | Extended thinking |-| **Google** | Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 2.0 Flash Thinking | Multimodal, 1M context |+| **OpenAI** | GPT-5, GPT-4o, o1, o3, o4 | 128K output, reasoning effort |+| **Anthropic** | Claude Opus 4.6, Sonnet 4.5 | 128K output, extended thinking |+| **Google** | Gemini 3.0, 2.5 Pro/Flash | 65K output, 1M context || **OpenRouter** | 200+ models | Unified API for all providers |-| **Groq** | Llama, Mixtral, Qwen | Ultra-fast inference |+| **Groq** | Llama, Qwen | Ultra-fast inference || **Cerebras** | Llama, Qwen | Ultra-fast on custom hardware |### AI Agents & Skillsdiff --git a/client/src/components/AIAgentNode.tsx b/client/src/components/AIAgentNode.tsxindex 2d9b0d45..5c347498 100644--- a/client/src/components/AIAgentNode.tsx+++ b/client/src/components/AIAgentNode.tsx@@ -66,28 +66,29 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},chatAgent: {icon: <span style={{ fontSize: '28px' }}>🧞</span>,title: 'Zeenie',- subtitle: 'Your Personal Assistant',+ subtitle: 'Personal Assistant',themeColorKey: 'cyan',bottomHandles: [{ id: 'input-skill', label: 'Skill', position: '25%' },{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],+ topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},socialReceive: {icon: <SocialReceiveIcon />,@@ -122,7 +123,7 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {},android_agent: {icon: <span style={{ fontSize: '28px' }}>📱</span>,- title: 'Android Control Agent',+ title: 'Android Agent',subtitle: 'Device Control',themeColorKey: 'green',bottomHandles: [@@ -130,12 +131,12 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},coding_agent: {icon: <span style={{ fontSize: '28px' }}>💻</span>,@@ -147,16 +148,16 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},web_agent: {icon: <span style={{ fontSize: '28px' }}>🌐</span>,- title: 'Web Control Agent',+ title: 'Web Agent',subtitle: 'Browser Automation',themeColorKey: 'pink',bottomHandles: [@@ -164,16 +165,16 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},task_agent: {icon: <span style={{ fontSize: '28px' }}>📋</span>,- title: 'Task Management Agent',+ title: 'Task Agent',subtitle: 'Task Automation',themeColorKey: 'purple',bottomHandles: [@@ -181,16 +182,16 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},social_agent: {icon: <span style={{ fontSize: '28px' }}>📱</span>,- title: 'Social Media Agent',+ title: 'Social Agent',subtitle: 'Social Messaging',themeColorKey: 'green',bottomHandles: [@@ -198,12 +199,12 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},travel_agent: {icon: <span style={{ fontSize: '28px' }}>✈️</span>,@@ -215,12 +216,12 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},tool_agent: {icon: <span style={{ fontSize: '28px' }}>🔧</span>,@@ -232,29 +233,29 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},productivity_agent: {icon: <span style={{ fontSize: '28px' }}>⏰</span>,title: 'Productivity Agent',- subtitle: 'Productivity',+ subtitle: 'Workflows',themeColorKey: 'cyan',bottomHandles: [{ id: 'input-skill', label: 'Skill', position: '25%' },{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},payments_agent: {icon: <span style={{ fontSize: '28px' }}>💳</span>,@@ -266,12 +267,12 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,},consumer_agent: {icon: <span style={{ fontSize: '28px' }}>🛒</span>,@@ -283,12 +284,46 @@ const AGENT_CONFIGS: Record<string, AgentConfig> = {{ id: 'input-tools', label: 'Tool', position: '75%' },],leftHandles: [- { id: 'input-memory', label: 'Memory', position: '55%' },+ { id: 'input-memory', label: 'Memory', position: '65%' },{ id: 'input-task', label: 'Task', position: '85%' },],topOutputHandle: { id: 'output-top', label: 'Output' },width: 260,- height: 160,+ height: 200,+ },+ autonomous_agent: {+ icon: <span style={{ fontSize: '28px' }}>🎯</span>,+ title: 'Autonomous Agent',+ subtitle: 'Autonomous Ops',+ themeColorKey: 'purple',+ bottomHandles: [+ { id: 'input-skill', label: 'Skill', position: '25%' },+ { id: 'input-tools', label: 'Tool', position: '75%' },+ ],+ leftHandles: [+ { id: 'input-memory', label: 'Memory', position: '65%' },+ { id: 'input-task', label: 'Task', position: '85%' },+ ],+ topOutputHandle: { id: 'output-top', label: 'Output' },+ width: 260,+ height: 200,+ },+ orchestrator_agent: {+ icon: <span style={{ fontSize: '28px' }}>🎼</span>,+ title: 'Orchestrator Agent',+ subtitle: 'Agent Coordination',+ themeColorKey: 'cyan',+ bottomHandles: [+ { id: 'input-skill', label: 'Skill', position: '25%' },+ { id: 'input-tools', label: 'Tool', position: '75%' },+ ],+ leftHandles: [+ { id: 'input-memory', label: 'Memory', position: '65%' },+ { id: 'input-task', label: 'Task', position: '85%' },+ ],+ topOutputHandle: { id: 'output-top', label: 'Output' },+ width: 260,+ height: 200,},};diff --git a/client/src/components/CredentialsModal.tsx b/client/src/components/CredentialsModal.tsxindex b8700c7a..041cdd13 100644--- a/client/src/components/CredentialsModal.tsx+++ b/client/src/components/CredentialsModal.tsx@@ -13,7 +13,7 @@ import {import Modal from './ui/Modal';import QRCodeDisplay from './ui/QRCodeDisplay';import ApiKeyInput from './ui/ApiKeyInput';-import { useApiKeys } from '../hooks/useApiKeys';+import { useApiKeys, ProviderDefaults } from '../hooks/useApiKeys';import { useAppTheme } from '../hooks/useAppTheme';import { useWhatsAppStatus, useAndroidStatus, useWebSocket, RateLimitConfig, RateLimitStats } from '../contexts/WebSocketContext';import { useWhatsApp } from '../hooks/useWhatsApp';@@ -114,7 +114,7 @@ interface Props {const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {const theme = useAppTheme();- const { validateApiKey, saveApiKey, getStoredApiKey, hasStoredKey, removeApiKey, validateGoogleMapsKey, isConnected } = useApiKeys();+ const { validateApiKey, saveApiKey, getStoredApiKey, hasStoredKey, removeApiKey, validateGoogleMapsKey, getProviderDefaults, saveProviderDefaults, isConnected } = useApiKeys();const whatsappStatus = useWhatsAppStatus();const androidStatus = useAndroidStatus();@@ -138,6 +138,11 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {const [proxyUrls, setProxyUrls] = useState<Record<string, string>>({});const [selectedItem, setSelectedItem] = useState<CredentialItem | null>(null);+ // Provider defaults state+ const [providerDefaults, setProviderDefaults] = useState<Record<string, ProviderDefaults>>({});+ const [defaultsLoading, setDefaultsLoading] = useState<Record<string, boolean>>({});+ const [defaultsDirty, setDefaultsDirty] = useState<Record<string, boolean>>({});+// Load stored keys and proxy URLsconst loadKeys = useCallback(async () => {const allItems = CATEGORIES.flatMap(c => c.items);@@ -177,6 +182,42 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {}}, [visible, isConnected, loadKeys]);+ // Load provider defaults when AI provider is selected+ useEffect(() => {+ const isAIProvider = CATEGORIES.find(c => c.key === 'ai')?.items.some(i => i.id === selectedItem?.id);+ if (selectedItem && !selectedItem.isSpecial && isAIProvider && isConnected) {+ // Only load if we don't already have defaults for this provider+ if (!providerDefaults[selectedItem.id]) {+ const loadDefaults = async () => {+ setDefaultsLoading(l => ({ ...l, [selectedItem.id]: true }));+ const defaults = await getProviderDefaults(selectedItem.id);+ setProviderDefaults(p => ({ ...p, [selectedItem.id]: defaults }));+ setDefaultsLoading(l => ({ ...l, [selectedItem.id]: false }));+ };+ loadDefaults();+ }+ }+ }, [selectedItem, isConnected, getProviderDefaults, providerDefaults]);++ // Handler to update a single default value+ const updateProviderDefault = (provider: string, key: keyof ProviderDefaults, value: number | boolean | string) => {+ setProviderDefaults(p => ({+ ...p,+ [provider]: { ...(p[provider] || {}), [key]: value } as ProviderDefaults+ }));+ setDefaultsDirty(d => ({ ...d, [provider]: true }));+ };++ // Handler to save provider defaults+ const handleSaveProviderDefaults = async (provider: string) => {+ setDefaultsLoading(l => ({ ...l, [provider]: true }));+ const success = await saveProviderDefaults(provider, providerDefaults[provider]);+ if (success) {+ setDefaultsDirty(d => ({ ...d, [provider]: false }));+ }+ setDefaultsLoading(l => ({ ...l, [provider]: false }));+ };+const handleValidate = async (id: string) => {const key = keys[id];if (!key?.trim()) return;@@ -323,7 +364,7 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {// WhatsApp actionsconst { getStatus: getWhatsAppStatus, startConnection, restartConnection } = useWhatsApp();- const { sendRequest, getWhatsAppRateLimitConfig, setWhatsAppRateLimitConfig, unpauseWhatsAppRateLimit } = useWebSocket();+ const { sendRequest, setAndroidStatus, getWhatsAppRateLimitConfig, setWhatsAppRateLimitConfig, unpauseWhatsAppRateLimit } = useWebSocket();// Android stateconst [androidApiKey, setAndroidApiKey] = useState('');@@ -381,8 +422,19 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {url: import.meta.env.VITE_ANDROID_RELAY_URL || '',api_key: key,});+ console.log('[Android] Connect response:', response);if (!response.success) {setAndroidError(response.error || 'Failed to connect');+ } else if (response.qr_data) {+ console.log('[Android] Setting QR data from response');+ // Use response data directly - don't rely on broadcast timing+ setAndroidStatus(prev => ({+ ...prev,+ connected: true,+ paired: false,+ qr_data: response.qr_data,+ session_token: response.session_token || prev.session_token,+ }));}} catch (err: any) {setAndroidError(err.message || 'Failed to connect');@@ -592,7 +644,7 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {<QRCodeDisplayvalue={whatsappStatus.qr}isConnected={whatsappStatus.connected}- size={200}+ size={280}connectedTitle="Already Connected!"connectedSubtitle="No QR code needed"loading={whatsappStatus.running && !whatsappStatus.qr && !whatsappStatus.connected}@@ -916,12 +968,12 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {backgroundColor: theme.colors.backgroundAlt,borderRadius: theme.borderRadius.lg,marginBottom: theme.spacing.xl,- minHeight: 200,+ minHeight: 300,}}><QRCodeDisplayvalue={androidStatus.qr_data}isConnected={androidStatus.paired}- size={260}+ size={280}connectedTitle="Device Paired!"connectedSubtitle={androidStatus.device_name || androidStatus.device_id || 'Android Device'}loading={androidStatus.connected && !androidStatus.qr_data && !androidStatus.paired}@@ -1213,6 +1265,110 @@ const CredentialsModal: React.FC<Props> = ({ visible, onClose }) => {</div>)}+ {/* Default Parameters Section - Only for AI providers */}+ {CATEGORIES.find(c => c.key === 'ai')?.items.some(i => i.id === item.id) && (+ <div style={{ marginBottom: theme.spacing.xl }}>+ <label style={{+ display: 'block',+ fontSize: theme.fontSize.sm,+ fontWeight: theme.fontWeight.medium,+ color: theme.colors.text,+ marginBottom: theme.spacing.md,+ }}>+ Default Parameters+ </label>++ <div style={{+ padding: theme.spacing.lg,+ backgroundColor: theme.colors.backgroundAlt,+ borderRadius: theme.borderRadius.md,+ border: `1px solid ${theme.colors.border}`,+ opacity: defaultsLoading[item.id] ? 0.6 : 1,+ }}>+ {/* Temperature */}+ <div style={{ display: 'flex', alignItems: 'center', justifyContent: 'space-between', marginBottom: theme.spacing.md }}>+ <div>+ <div style={{ fontSize: theme.fontSize.sm, fontWeight: theme.fontWeight.medium, color: theme.colors.text }}>Temperature</div>+ <div style={{ fontSize: theme.fontSize.xs, color: theme.colors.textSecondary }}>Controls randomness (0-2)</div>+ </div>+ <InputNumber+ size="small"+ value={providerDefaults[item.id]?.temperature ?? 0.7}+ onChange={(v) => updateProviderDefault(item.id, 'temperature', v ?? 0.7)}+ min={0}+ max={2}+ step={0.1}+ style={{ width: 80 }}+ />+ </div>++ {/* Max Tokens */}+ <div style={{ display: 'flex', alignItems: 'center', justifyContent: 'space-between', marginBottom: theme.spacing.md }}>+ <div>+ <div style={{ fontSize: theme.fontSize.sm, fontWeight: theme.fontWeight.medium, color: theme.colors.text }}>Max Tokens</div>+ <div style={{ fontSize: theme.fontSize.xs, color: theme.colors.textSecondary }}>Maximum response length</div>+ </div>+ <InputNumber+ size="small"+ value={providerDefaults[item.id]?.max_tokens ?? 4096}+ onChange={(v) => updateProviderDefault(item.id, 'max_tokens', v ?? 4096)}+ min={1}+ max={128000}+ style={{ width: 100 }}+ />+ </div>++ {/* Thinking/Reasoning Toggle */}+ <div style={{ display: 'flex', alignItems: 'center', justifyContent: 'space-between', marginBottom: theme.spacing.md }}>+ <div>+ <div style={{ fontSize: theme.fontSize.sm, fontWeight: theme.fontWeight.medium, color: theme.colors.text }}>Thinking/Reasoning</div>+ <div style={{ fontSize: theme.fontSize.xs, color: theme.colors.textSecondary }}>Extended thinking for supported models</div>+ </div>+ <Switch+ size="small"+ checked={providerDefaults[item.id]?.thinking_enabled ?? false}+ onChange={(v) => updateProviderDefault(item.id, 'thinking_enabled', v)}+ />+ </div>++ {/* Thinking Budget - conditional */}+ {providerDefaults[item.id]?.thinking_enabled && (+ <div style={{ display: 'flex', alignItems: 'center', justifyContent: 'space-between', marginBottom: theme.spacing.md }}>+ <div>+ <div style={{ fontSize: theme.fontSize.sm, fontWeight: theme.fontWeight.medium, color: theme.colors.text }}>Thinking Budget</div>+ <div style={{ fontSize: theme.fontSize.xs, color: theme.colors.textSecondary }}>Token budget (1024-16000)</div>+ </div>+ <InputNumber+ size="small"+ value={providerDefaults[item.id]?.thinking_budget ?? 2048}+ onChange={(v) => updateProviderDefault(item.id, 'thinking_budget', v ?? 2048)}+ min={1024}+ max={16000}+ style={{ width: 80 }}+ />+ </div>+ )}++ {/* Save Button */}+ <Button+ size="small"+ onClick={() => handleSaveProviderDefaults(item.id)}+ loading={defaultsLoading[item.id]}+ disabled={!defaultsDirty[item.id]}+ block+ style={{+ marginTop: theme.spacing.sm,+ backgroundColor: defaultsDirty[item.id] ? `${theme.dracula.green}25` : undefined,+ borderColor: defaultsDirty[item.id] ? `${theme.dracula.green}60` : undefined,+ color: defaultsDirty[item.id] ? theme.dracula.green : undefined,+ }}+ >+ Save Defaults+ </Button>+ </div>+ </div>+ )}+{/* Models list */}{models[item.id]?.length > 0 && (<div style={{ marginBottom: theme.spacing.xl }}>diff --git a/client/src/components/ParameterRenderer.tsx b/client/src/components/ParameterRenderer.tsxindex 3d0f6ce5..27e29d8e 100644--- a/client/src/components/ParameterRenderer.tsx+++ b/client/src/components/ParameterRenderer.tsx@@ -768,6 +768,8 @@ interface ParameterRendererProps {onParameterChange?: (paramName: string, value: any) => void;onClosePanel?: () => void;isLoadingParameters?: boolean;+ /** For memory nodes: the connected agent's node ID for auto-session display */+ connectedAgentId?: string | null;}// Type guard to check if parameter is INodeProperties@@ -782,6 +784,7 @@ const ParameterRenderer: React.FC<ParameterRendererProps> = ({allParameters,onParameterChange,isLoadingParameters = false,+ connectedAgentId,}) => {const theme = useAppTheme();// Don't use default while loading - wait for actual saved value to load@@ -1380,6 +1383,44 @@ const ParameterRenderer: React.FC<ParameterRendererProps> = ({);}+ // Special case for sessionId parameter - show auto-derived session ID in placeholder+ if (parameter.name === 'sessionId' && connectedAgentId) {+ const effectivePlaceholder = connectedAgentId;++ return (+ <input+ type="text"+ value={currentValue || ''}+ onChange={(e) => onChange(e.target.value)}+ placeholder={effectivePlaceholder}+ onDragOver={handleDragOver}+ onDragLeave={handleDragLeave}+ onDrop={handleDrop}+ style={{+ width: '100%',+ padding: '10px 12px',+ border: isDragOver ? `2px solid ${theme.accent.cyan}` : `1px solid ${theme.colors.border}`,+ borderRadius: '6px',+ fontSize: '14px',+ backgroundColor: isDragOver ? `${theme.accent.cyan}10` : theme.colors.backgroundAlt,+ color: currentValue && currentValue.includes('{{') ? theme.accent.yellow : theme.colors.text,+ outline: 'none',+ transition: 'all 0.2s ease',+ fontFamily: currentValue && currentValue.includes('{{') ? 'monospace' : 'system-ui, sans-serif',+ boxShadow: 'inset 0 1px 2px rgba(0,0,0,0.05)'+ }}+ onFocus={(e) => {+ e.target.style.borderColor = theme.accent.cyan;+ e.target.style.boxShadow = `0 0 0 3px ${theme.accent.cyan}20, inset 0 1px 2px rgba(0,0,0,0.05)`;+ }}+ onBlur={(e) => {+ e.target.style.borderColor = theme.colors.border;+ e.target.style.boxShadow = 'inset 0 1px 2px rgba(0,0,0,0.05)';+ }}+ />+ );+ }+return (<inputtype={isPassword ? "password" : "text"}diff --git a/client/src/components/parameterPanel/InputSection.tsx b/client/src/components/parameterPanel/InputSection.tsxindex 01e36b06..65f2fb7c 100644--- a/client/src/components/parameterPanel/InputSection.tsx+++ b/client/src/components/parameterPanel/InputSection.tsx@@ -88,7 +88,8 @@ const InputSection: React.FC<InputSectionProps> = ({ nodeId, visible = true }) =const AGENT_WITH_SKILLS_TYPES = ['aiAgent', 'chatAgent', 'android_agent', 'coding_agent','web_agent', 'task_agent', 'social_agent', 'travel_agent',- 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent'+ 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent',+ 'autonomous_agent', 'orchestrator_agent'];// Check if current node is an agent that shows skills in Middle Sectionconst isAgentWithSkills = AGENT_WITH_SKILLS_TYPES.includes(currentNodeType || '');@@ -486,7 +487,7 @@ const InputSection: React.FC<InputSectionProps> = ({ nodeId, visible = true }) =const isMemory = nodeType === 'simpleMemory';const nodeTypeLower = nodeType.toLowerCase();// AI agent types that use the AI output schema- const aiAgentTypes = ['aiAgent', 'chatAgent', 'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent'];+ const aiAgentTypes = ['aiAgent', 'chatAgent', 'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent', 'autonomous_agent', 'orchestrator_agent'];const isAI = !isMemory && (nodeTypeLower.includes('chatmodel') || aiAgentTypes.includes(nodeType));const isFile = nodeType.includes('file');const isWhatsAppDb = nodeType === 'whatsappDb';diff --git a/client/src/components/parameterPanel/MasterSkillEditor.tsx b/client/src/components/parameterPanel/MasterSkillEditor.tsxindex cb237c77..f52dbffa 100644--- a/client/src/components/parameterPanel/MasterSkillEditor.tsx+++ b/client/src/components/parameterPanel/MasterSkillEditor.tsx@@ -6,11 +6,13 @@** Skills loaded from skillFolder (server/skills/<folder>/) or built-in list.* The skillsConfig uses skillName (folder name) as keys.+ *+ * User skills are created/edited inline in the right panel (no modal).*/import React, { useState, useEffect, useCallback, useMemo, useRef } from 'react';-import { Input, Select, Checkbox, Button, Spin, List, Badge, Empty, Tooltip, Alert } from 'antd';-import { SearchOutlined, ReloadOutlined, InfoCircleOutlined, FolderOutlined } from '@ant-design/icons';+import { Input, Select, Checkbox, Button, Spin, List, Badge, Empty, Tooltip, Alert, Popconfirm, message } from 'antd';+import { SearchOutlined, ReloadOutlined, InfoCircleOutlined, FolderOutlined, PlusOutlined, DeleteOutlined, SaveOutlined, CloseOutlined } from '@ant-design/icons';import { useAppTheme } from '../../hooks/useAppTheme';import { useWebSocket } from '../../contexts/WebSocketContext';import { skillNodes, SKILL_NODE_TYPES } from '../../nodeDefinitions/skillNodes';@@ -34,6 +36,29 @@ interface AvailableSkill {icon: string;color: string;description: string;+ isUserSkill?: boolean; // True if this is a user-created skill from database+}++// User skill from database+interface UserSkill {+ name: string;+ display_name: string;+ description: string;+ instructions: string;+ icon: string;+ color: string;+ category: string;+ is_active: boolean;+}++// Pending skill data for create/edit+interface PendingSkillData {+ name: string;+ display_name: string;+ description: string;+ instructions: string;+ icon: string;+ color: string;}interface MasterSkillEditorProps {@@ -41,6 +66,7 @@ interface MasterSkillEditorProps {onConfigChange: (config: MasterSkillConfig) => void;skillFolder?: string;onSkillFolderChange?: (folder: string) => void;+ nodeId?: string; // For persisting skillsConfig to database}// Lookup icon/color from skillNodes.ts definitions by skill name@@ -71,7 +97,8 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({skillsConfig,onConfigChange,skillFolder,- onSkillFolderChange+ onSkillFolderChange,+ nodeId}) => {const theme = useAppTheme();const { sendRequest } = useWebSocket();@@ -86,9 +113,17 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({const hasFetchedFolders = useRef(false);const editorWrapperRef = useRef<HTMLDivElement>(null);+ // User skills from database+ const [userSkills, setUserSkills] = useState<UserSkill[]>([]);++ // Inline editing state (no modal)+ const [isCreatingNew, setIsCreatingNew] = useState(false);+ const [pendingSkillData, setPendingSkillData] = useState<PendingSkillData | null>(null);+ const [savingSkill, setSavingSkill] = useState(false);+ const [hasUnsavedChanges, setHasUnsavedChanges] = useState(false);+// Native DOM keydown handler to stop React Flow's document-level listeners// from intercepting standard text editing shortcuts (Ctrl+A, Ctrl+C, etc.)- // React synthetic stopPropagation doesn't prevent native document.addEventListener handlers.useEffect(() => {const el = editorWrapperRef.current;if (!el) return;@@ -127,6 +162,22 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({fetchFolders();}, [sendRequest]);+ // Fetch user-created skills from database+ const fetchUserSkills = useCallback(async () => {+ try {+ const response = await sendRequest<{ skills: UserSkill[]; count: number }>('get_user_skills', { active_only: false });+ if (response?.skills) {+ setUserSkills(response.skills);+ }+ } catch (error) {+ console.error('[MasterSkillEditor] Failed to fetch user skills:', error);+ }+ }, [sendRequest]);++ useEffect(() => {+ fetchUserSkills();+ }, [fetchUserSkills]);+// Fetch skills from folder when skillFolder is setuseEffect(() => {if (!skillFolder) {@@ -150,7 +201,7 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({type: s.name,skillName: s.name,displayName: s.name.split('-').map(w => w[0].toUpperCase() + w.slice(1)).join(' '),- icon: defaults.icon || s.metadata?.icon || '📄',+ icon: defaults.icon || s.metadata?.icon || '',color: defaults.color || s.metadata?.color || '#6366F1',description: s.description || ''};@@ -166,32 +217,50 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({fetchFolderSkills();}, [skillFolder, sendRequest]);- // Build list of available skills - from folder scan or node definitions+ // Build list of available skills - from folder scan, node definitions, and user skillsconst availableSkills = useMemo<AvailableSkill[]>(() => {+ const skills: AvailableSkill[] = [];++ // Add folder skills if availableif (skillFolder && folderSkills.length > 0) {- return folderSkills;+ skills.push(...folderSkills);+ } else {+ // Add built-in skills from node definitions+ SKILL_NODE_TYPES+ .filter(type => type !== 'customSkill' && type !== 'masterSkill')+ .forEach(type => {+ const nodeDef = skillNodes[type];+ if (!nodeDef) return;++ const skillNameProp = nodeDef.properties?.find(p => p.name === 'skillName');+ const skillName = skillNameProp?.default as string || type;++ skills.push({+ type,+ skillName,+ displayName: nodeDef.displayName || type,+ icon: nodeDef.icon || '',+ color: (nodeDef.defaults?.color as string) || '#6366F1',+ description: nodeDef.description || ''+ });+ });}- return SKILL_NODE_TYPES- .filter(type => type !== 'customSkill' && type !== 'masterSkill')- .map(type => {- const nodeDef = skillNodes[type];- if (!nodeDef) return null;-- const skillNameProp = nodeDef.properties?.find(p => p.name === 'skillName');- const skillName = skillNameProp?.default as string || type;-- return {- type,- skillName, // This is the folder name used as config key- displayName: nodeDef.displayName || type,- icon: nodeDef.icon || '',- color: (nodeDef.defaults?.color as string) || '#6366F1',- description: nodeDef.description || ''- };- })- .filter((s): s is AvailableSkill => s !== null);- }, [skillFolder, folderSkills]);+ // Add user-created skills+ userSkills.forEach(us => {+ skills.push({+ type: 'userSkill',+ skillName: us.name,+ displayName: us.display_name,+ icon: us.icon || '',+ color: us.color || '#6366F1',+ description: us.description || '',+ isUserSkill: true+ });+ });++ return skills;+ }, [skillFolder, folderSkills, userSkills]);// Filter skills based on search queryconst filteredSkills = useMemo(() => {@@ -203,12 +272,34 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({);}, [availableSkills, searchQuery]);+ // Clean up skillsConfig - remove skills that no longer exist in availableSkills+ useEffect(() => {+ if (availableSkills.length === 0) return; // Wait for skills to load++ const availableSkillNames = new Set(availableSkills.map(s => s.skillName));+ const configSkillNames = Object.keys(skillsConfig);+ const staleSkills = configSkillNames.filter(name => !availableSkillNames.has(name));++ if (staleSkills.length > 0) {+ console.log('[MasterSkillEditor] Removing stale skills from config:', staleSkills);+ const cleanedConfig = { ...skillsConfig };+ staleSkills.forEach(name => delete cleanedConfig[name]);+ onConfigChange(cleanedConfig);++ // Clear selection if selected skill was stale+ if (selectedSkillName && staleSkills.includes(selectedSkillName)) {+ setSelectedSkillName(null);+ setPendingSkillData(null);+ }+ }+ }, [availableSkills, skillsConfig, onConfigChange, selectedSkillName]);+// Auto-select first skilluseEffect(() => {- if (!selectedSkillName && availableSkills.length > 0) {+ if (!selectedSkillName && !isCreatingNew && availableSkills.length > 0) {setSelectedSkillName(availableSkills[0].skillName);}- }, [selectedSkillName, availableSkills]);+ }, [selectedSkillName, isCreatingNew, availableSkills]);// Load skill content from skill folderconst fetchSkillContent = useCallback(async (skillName: string): Promise<string> => {@@ -239,14 +330,44 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({// Load instructions when skill is selected (for preview)useEffect(() => {- if (!selectedSkillName) return;+ if (!selectedSkillName || isCreatingNew) return;++ // Skip if skill doesn't exist in availableSkills (may have been deleted)+ const skillExists = availableSkills.some(s => s.skillName === selectedSkillName);+ if (!skillExists) return;// Skip if already cachedif (defaultInstructions[selectedSkillName]) return;// Load skill content for previewfetchSkillContent(selectedSkillName);- }, [selectedSkillName, defaultInstructions, fetchSkillContent]);+ }, [selectedSkillName, isCreatingNew, defaultInstructions, fetchSkillContent, availableSkills]);++ // When selecting a user skill, load its data into pendingSkillData for editing+ useEffect(() => {+ if (isCreatingNew || !selectedSkillName) {+ return;+ }++ const selectedInfo = availableSkills.find(s => s.skillName === selectedSkillName);+ if (selectedInfo?.isUserSkill) {+ const userSkill = userSkills.find(us => us.name === selectedSkillName);+ if (userSkill) {+ setPendingSkillData({+ name: userSkill.name,+ display_name: userSkill.display_name,+ description: userSkill.description,+ instructions: userSkill.instructions,+ icon: userSkill.icon || '',+ color: userSkill.color || '#6366F1'+ });+ setHasUnsavedChanges(false);+ }+ } else {+ setPendingSkillData(null);+ setHasUnsavedChanges(false);+ }+ }, [selectedSkillName, isCreatingNew, availableSkills, userSkills]);// Toggle skill enabled/disabledconst handleToggleSkill = useCallback(async (skillName: string, enabled: boolean) => {@@ -271,7 +392,7 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({}}, [skillsConfig, onConfigChange, fetchSkillContent]);- // Update instructions+ // Update instructions (for built-in skills)const handleUpdateInstructions = useCallback((skillName: string, instructions: string) => {const currentConfig = skillsConfig[skillName];const defaultContent = defaultInstructions[skillName] || '';@@ -305,11 +426,145 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({});}, [skillsConfig, onConfigChange, fetchSkillContent]);+ // Create new skill - show inline editor+ const handleCreateSkill = useCallback(() => {+ setIsCreatingNew(true);+ setSelectedSkillName(null);+ setPendingSkillData({+ name: '',+ display_name: '',+ description: '',+ instructions: '# Skill Instructions\n\nDescribe what this skill does and how the AI should use it.',+ icon: '',+ color: '#6366F1'+ });+ setHasUnsavedChanges(false);+ }, []);++ // Cancel creating new skill+ const handleCancelCreate = useCallback(() => {+ setIsCreatingNew(false);+ setPendingSkillData(null);+ setHasUnsavedChanges(false);+ // Select first available skill+ if (availableSkills.length > 0) {+ setSelectedSkillName(availableSkills[0].skillName);+ }+ }, [availableSkills]);++ // Update pending skill data+ const handlePendingDataChange = useCallback((field: keyof PendingSkillData, value: string) => {+ setPendingSkillData(prev => prev ? { ...prev, [field]: value } : null);+ setHasUnsavedChanges(true);+ }, []);++ // Save skill (create or update)+ const handleSaveSkill = useCallback(async () => {+ if (!pendingSkillData) return;++ // Validate required fields+ if (!pendingSkillData.display_name.trim()) {+ message.error('Display name is required');+ return;+ }+ if (!pendingSkillData.instructions.trim()) {+ message.error('Instructions are required');+ return;+ }++ // For new skills, generate name from display_name if not provided+ let skillName = pendingSkillData.name;+ if (isCreatingNew) {+ if (!skillName.trim()) {+ skillName = pendingSkillData.display_name+ .toLowerCase()+ .replace(/[^a-z0-9]+/g, '-')+ .replace(/^-|-$/g, '');+ }+ if (!skillName || !/^[a-z0-9-]+$/.test(skillName)) {+ message.error('Invalid skill ID. Use lowercase letters, numbers, and hyphens only.');+ return;+ }+ }++ setSavingSkill(true);+ try {+ const handler = isCreatingNew ? 'create_user_skill' : 'update_user_skill';+ const payload = {+ name: skillName,+ display_name: pendingSkillData.display_name,+ description: pendingSkillData.description,+ instructions: pendingSkillData.instructions,+ icon: pendingSkillData.icon,+ color: pendingSkillData.color,+ category: skillFolder || 'custom',+ is_active: true+ };++ const result = await sendRequest<{ skill?: UserSkill; success?: boolean; error?: string }>(handler, payload);++ if (result.skill || result.success) {+ message.success(isCreatingNew ? 'Skill created' : 'Skill saved');+ await fetchUserSkills();++ if (isCreatingNew) {+ // Add to config as enabled+ onConfigChange({+ ...skillsConfig,+ [skillName]: { enabled: true, instructions: pendingSkillData.instructions, isCustomized: false }+ });+ setIsCreatingNew(false);+ setSelectedSkillName(skillName);+ }+ setHasUnsavedChanges(false);+ } else {+ message.error(result.error || 'Failed to save skill');+ }+ } catch (err: any) {+ message.error(err.message || 'Failed to save skill');+ } finally {+ setSavingSkill(false);+ }+ }, [pendingSkillData, isCreatingNew, skillFolder, sendRequest, fetchUserSkills, skillsConfig, onConfigChange]);++ // Delete user skill+ const handleDeleteSkill = useCallback(async (skillName: string) => {+ try {+ const result = await sendRequest<{ success: boolean; error?: string }>('delete_user_skill', { name: skillName });+ if (result.success) {+ message.success('Skill deleted');+ // Remove from config+ const newConfig = { ...skillsConfig };+ delete newConfig[skillName];+ onConfigChange(newConfig);++ // Persist cleaned config to database so deleted skill doesn't reappear+ if (nodeId) {+ await sendRequest('save_node_parameters', {+ node_id: nodeId,+ parameters: { skillsConfig: newConfig, skillFolder: skillFolder || 'assistant' }+ });+ }++ // Clear selection if deleted skill was selected+ if (selectedSkillName === skillName) {+ setSelectedSkillName(null);+ setPendingSkillData(null);+ }+ // Refresh user skills list from database+ await fetchUserSkills();+ } else {+ message.error(result.error || 'Failed to delete skill');+ }+ } catch (err: any) {+ message.error(err.message || 'Failed to delete skill');+ }+ }, [sendRequest, fetchUserSkills, skillsConfig, onConfigChange, selectedSkillName, nodeId, skillFolder]);+const selectedSkillInfo = availableSkills.find(s => s.skillName === selectedSkillName);const selectedSkillConfig = selectedSkillName ? skillsConfig[selectedSkillName] : undefined;const enabledCount = Object.values(skillsConfig).filter(c => c?.enabled).length;-- console.log('[MasterSkillEditor] RENDER', { skillFolder, onSkillFolderChange: !!onSkillFolderChange, skillsConfig: Object.keys(skillsConfig) });+ const isEditingUserSkill = selectedSkillInfo?.isUserSkill && pendingSkillData;return (<div style={{@@ -329,7 +584,7 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({border: `1px solid ${theme.colors.border}`,overflow: 'hidden'}}>- {/* Header with count */}+ {/* Header with count and create button */}<div style={{padding: theme.spacing.sm,borderBottom: `1px solid ${theme.colors.border}`,@@ -341,11 +596,22 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({<span style={{ fontSize: theme.fontSize.sm, fontWeight: theme.fontWeight.semibold, color: theme.colors.text }}>Skills</span>- <Badge- count={enabledCount}- style={{ backgroundColor: theme.dracula.purple }}- showZero- />+ <div style={{ display: 'flex', alignItems: 'center', gap: theme.spacing.xs }}>+ <Badge count={enabledCount} style={{ backgroundColor: theme.dracula.purple }} showZero />+ <Tooltip title="Create new skill">+ <Button+ size="small"+ icon={<PlusOutlined />}+ onClick={handleCreateSkill}+ disabled={isCreatingNew}+ style={{+ backgroundColor: `${theme.dracula.green}20`,+ borderColor: theme.dracula.green,+ color: theme.dracula.green,+ }}+ />+ </Tooltip>+ </div></div>{/* Search */}@@ -401,13 +667,22 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({dataSource={filteredSkills}renderItem={(skill) => {const config = skillsConfig[skill.skillName];- const isSelected = selectedSkillName === skill.skillName;+ const isSelected = selectedSkillName === skill.skillName && !isCreatingNew;const isEnabled = config?.enabled || false;const isCustomized = config?.isCustomized || false;return (<List.Item- onClick={() => setSelectedSkillName(skill.skillName)}+ onClick={() => {+ if (isCreatingNew) {+ // Confirm cancel if creating new+ if (hasUnsavedChanges) {+ if (!confirm('Discard unsaved changes?')) return;+ }+ setIsCreatingNew(false);+ }+ setSelectedSkillName(skill.skillName);+ }}style={{padding: `${theme.spacing.sm} ${theme.spacing.md}`,cursor: 'pointer',@@ -425,7 +700,7 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({}}onClick={(e) => e.stopPropagation()}/>- {renderSkillIcon(skill.icon, 16)}+ {renderSkillIcon(skill.icon || '', 16)}<span style={{flex: 1,fontSize: theme.fontSize.sm,@@ -448,6 +723,19 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({}} /></Tooltip>)}+ {skill.isUserSkill && (+ <Badge+ count="Custom"+ style={{+ backgroundColor: `${theme.dracula.cyan}20`,+ color: theme.dracula.cyan,+ fontSize: 10,+ padding: '0 4px',+ height: 16,+ lineHeight: '16px'+ }}+ />+ )}</div></List.Item>);@@ -467,9 +755,10 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({border: `1px solid ${theme.colors.border}`,overflow: 'hidden'}}>- {selectedSkillInfo ? (+ {/* Creating new skill */}+ {isCreatingNew && pendingSkillData ? (<>- {/* Skill Header */}+ {/* New Skill Header */}<div style={{padding: theme.spacing.md,borderBottom: `1px solid ${theme.colors.border}`,@@ -478,35 +767,245 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({alignItems: 'center',gap: theme.spacing.md}}>- {renderSkillIcon(selectedSkillInfo.icon, 24)}+ <div style={{+ width: 40,+ height: 40,+ borderRadius: theme.borderRadius.md,+ backgroundColor: `${pendingSkillData.color}20`,+ display: 'flex',+ alignItems: 'center',+ justifyContent: 'center',+ border: `2px dashed ${pendingSkillData.color}`+ }}>+ {pendingSkillData.icon ? renderSkillIcon(pendingSkillData.icon, 20) : <PlusOutlined style={{ color: pendingSkillData.color }} />}+ </div><div style={{ flex: 1 }}><div style={{fontSize: theme.fontSize.sm,fontWeight: theme.fontWeight.semibold,- color: theme.colors.text,- display: 'flex',- alignItems: 'center',- gap: theme.spacing.sm+ color: theme.dracula.green}}>- {selectedSkillInfo.displayName}- {selectedSkillConfig?.isCustomized && (- <Badge- count="Customized"- style={{- backgroundColor: `${theme.dracula.orange}20`,- color: theme.dracula.orange,- fontSize: theme.fontSize.xs- }}- />- )}+ Create New Skill</div><div style={{ fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginTop: 2 }}>- {selectedSkillInfo.description}+ Fill in the details below and save+ </div>+ </div>+ <Button+ icon={<CloseOutlined />}+ size="small"+ onClick={handleCancelCreate}+ style={{+ backgroundColor: `${theme.dracula.red}15`,+ borderColor: `${theme.dracula.red}40`,+ color: theme.dracula.red,+ }}+ >+ Cancel+ </Button>+ <Button+ icon={<SaveOutlined />}+ size="small"+ onClick={handleSaveSkill}+ loading={savingSkill}+ style={{+ backgroundColor: `${theme.dracula.green}15`,+ borderColor: `${theme.dracula.green}40`,+ color: theme.dracula.green,+ }}+ >+ Save+ </Button>+ </div>++ {/* New Skill Form */}+ <div ref={editorWrapperRef} style={{ flex: 1, padding: theme.spacing.md, overflow: 'auto', display: 'flex', flexDirection: 'column', gap: theme.spacing.md }}>+ {/* Display Name */}+ <div>+ <label style={{ display: 'block', fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginBottom: 4 }}>+ Display Name *+ </label>+ <Input+ value={pendingSkillData.display_name}+ onChange={(e) => handlePendingDataChange('display_name', e.target.value)}+ placeholder="My Custom Skill"+ style={{ backgroundColor: theme.colors.backgroundAlt, borderColor: theme.colors.border }}+ />+ </div>++ {/* Description */}+ <div>+ <label style={{ display: 'block', fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginBottom: 4 }}>+ Description+ </label>+ <Input+ value={pendingSkillData.description}+ onChange={(e) => handlePendingDataChange('description', e.target.value)}+ placeholder="Brief description of what this skill does"+ style={{ backgroundColor: theme.colors.backgroundAlt, borderColor: theme.colors.border }}+ />+ </div>++ {/* Icon and Color */}+ <div style={{ display: 'flex', gap: theme.spacing.md }}>+ <div style={{ width: 100 }}>+ <label style={{ display: 'block', fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginBottom: 4 }}>+ Icon (emoji)+ </label>+ <Input+ value={pendingSkillData.icon}+ onChange={(e) => handlePendingDataChange('icon', e.target.value)}+ placeholder=""+ style={{ backgroundColor: theme.colors.backgroundAlt, borderColor: theme.colors.border, textAlign: 'center' }}+ />+ </div>+ <div style={{ width: 80 }}>+ <label style={{ display: 'block', fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginBottom: 4 }}>+ Color+ </label>+ <Input+ type="color"+ value={pendingSkillData.color}+ onChange={(e) => handlePendingDataChange('color', e.target.value)}+ style={{ backgroundColor: theme.colors.backgroundAlt, borderColor: theme.colors.border, height: 32, padding: 2 }}+ /></div></div>- {/* Reset Button */}- {selectedSkillConfig?.isCustomized && (+ {/* Instructions */}+ <div style={{ flex: 1, display: 'flex', flexDirection: 'column', minHeight: 200 }}>+ <label style={{ display: 'block', fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginBottom: 4 }}>+ Instructions *+ </label>+ <Input.TextArea+ value={pendingSkillData.instructions}+ onChange={(e) => handlePendingDataChange('instructions', e.target.value)}+ placeholder="# Skill Instructions..."+ spellCheck={false}+ autoSize={false}+ style={{+ flex: 1,+ minHeight: 0,+ resize: 'none',+ fontFamily: "'Consolas', 'Monaco', 'Fira Code', monospace",+ fontSize: 13,+ lineHeight: 1.5,+ backgroundColor: theme.colors.backgroundAlt,+ color: theme.colors.text,+ borderColor: theme.colors.border,+ borderRadius: theme.borderRadius.md,+ }}+ />+ </div>+ </div>+ </>+ ) : selectedSkillInfo ? (+ <>+ {/* Skill Header */}+ <div style={{+ padding: theme.spacing.md,+ borderBottom: `1px solid ${theme.colors.border}`,+ backgroundColor: theme.colors.backgroundAlt,+ display: 'flex',+ alignItems: 'center',+ gap: theme.spacing.md+ }}>+ {renderSkillIcon(selectedSkillInfo.icon || '', 24)}+ <div style={{ flex: 1 }}>+ {isEditingUserSkill ? (+ <Input+ value={pendingSkillData?.display_name || ''}+ onChange={(e) => handlePendingDataChange('display_name', e.target.value)}+ style={{+ fontSize: theme.fontSize.sm,+ fontWeight: theme.fontWeight.semibold,+ backgroundColor: theme.colors.background,+ borderColor: theme.colors.border,+ marginBottom: 4+ }}+ />+ ) : (+ <div style={{+ fontSize: theme.fontSize.sm,+ fontWeight: theme.fontWeight.semibold,+ color: theme.colors.text,+ display: 'flex',+ alignItems: 'center',+ gap: theme.spacing.sm+ }}>+ {selectedSkillInfo.displayName}+ {selectedSkillConfig?.isCustomized && (+ <Badge+ count="Customized"+ style={{+ backgroundColor: `${theme.dracula.orange}20`,+ color: theme.dracula.orange,+ fontSize: theme.fontSize.xs+ }}+ />+ )}+ </div>+ )}+ {isEditingUserSkill ? (+ <Input+ value={pendingSkillData?.description || ''}+ onChange={(e) => handlePendingDataChange('description', e.target.value)}+ placeholder="Description"+ style={{+ fontSize: theme.fontSize.xs,+ backgroundColor: theme.colors.background,+ borderColor: theme.colors.border,+ }}+ />+ ) : (+ <div style={{ fontSize: theme.fontSize.xs, color: theme.colors.textSecondary, marginTop: 2 }}>+ {selectedSkillInfo.description}+ </div>+ )}+ </div>++ {/* User skill actions */}+ {isEditingUserSkill && (+ <>+ {hasUnsavedChanges && (+ <Button+ icon={<SaveOutlined />}+ size="small"+ onClick={handleSaveSkill}+ loading={savingSkill}+ style={{+ backgroundColor: `${theme.dracula.green}15`,+ borderColor: `${theme.dracula.green}40`,+ color: theme.dracula.green,+ }}+ >+ Save+ </Button>+ )}+ <Popconfirm+ title="Delete this skill?"+ onConfirm={() => handleDeleteSkill(selectedSkillName!)}+ okText="Delete"+ cancelText="Cancel"+ okButtonProps={{ danger: true }}+ >+ <Button+ icon={<DeleteOutlined />}+ size="small"+ danger+ style={{+ backgroundColor: `${theme.dracula.red}15`,+ borderColor: `${theme.dracula.red}40`,+ }}+ >+ Delete+ </Button>+ </Popconfirm>+ </>+ )}++ {/* Reset Button for built-in skills */}+ {!isEditingUserSkill && selectedSkillConfig?.isCustomized && (<Buttonicon={<ReloadOutlined />}size="small"@@ -531,8 +1030,18 @@ const MasterSkillEditor: React.FC<MasterSkillEditorProps> = ({</div>) : (<Input.TextArea- value={selectedSkillConfig?.instructions || defaultInstructions[selectedSkillName!] || ''}- onChange={(e) => handleUpdateInstructions(selectedSkillName!, e.target.value)}+ value={+ isEditingUserSkill+ ? pendingSkillData?.instructions || ''+ : (selectedSkillConfig?.instructions || defaultInstructions[selectedSkillName!] || '')+ }+ onChange={(e) => {+ if (isEditingUserSkill) {+ handlePendingDataChange('instructions', e.target.value);+ } else {+ handleUpdateInstructions(selectedSkillName!, e.target.value);+ }+ }}placeholder="Loading skill instructions..."spellCheck={false}autoSize={false}diff --git a/client/src/components/parameterPanel/MiddleSection.tsx b/client/src/components/parameterPanel/MiddleSection.tsxindex f599bc1c..0a4a461b 100644--- a/client/src/components/parameterPanel/MiddleSection.tsx+++ b/client/src/components/parameterPanel/MiddleSection.tsx@@ -29,7 +29,9 @@ const AGENT_WITH_SKILLS_TYPES = ['tool_agent','productivity_agent','payments_agent',- 'consumer_agent'+ 'consumer_agent',+ 'autonomous_agent',+ 'orchestrator_agent'];interface ConnectedSkill {@@ -108,6 +110,9 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({const [editThresholdValue, setEditThresholdValue] = useState<number>(100000);const [savingThreshold, setSavingThreshold] = useState(false);+ // For Memory nodes: track the connected agent's ID for auto-session display+ const [connectedAgentId, setConnectedAgentId] = useState<string | null>(null);+// Clear memory handlerconst handleClearMemory = async () => {setIsProcessing(true);@@ -168,6 +173,26 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({// State for Master Skill parametersconst [masterSkillParams, setMasterSkillParams] = useState<Record<string, any>>({});+ // Load global memory window size setting for new Memory nodes+ useEffect(() => {+ if (!isMemoryNode) return;++ // Only apply global default if windowSize hasn't been explicitly set yet+ const loadGlobalWindowSize = async () => {+ try {+ const response = await sendRequest<{ settings: any }>('get_user_settings', {});+ const globalWindowSize = response?.settings?.memory_window_size;+ if (globalWindowSize !== undefined && parameters.windowSize === undefined) {+ onParameterChange('windowSize', globalWindowSize);+ }+ } catch (err) {+ console.error('[MiddleSection] Failed to load global memory settings:', err);+ }+ };++ loadGlobalWindowSize();+ }, [nodeId, isMemoryNode]); // Only run when node changes, not on every parameter change+// Get connected memory session ID and compaction stats for agent nodesuseEffect(() => {if (!isAgentWithSkills || !currentWorkflow) {@@ -190,13 +215,18 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({}// Get memory node's sessionId from its parameters+ // If empty or 'default', the actual session will be the agent's node_id (auto-derived)const fetchMemorySessionId = async () => {try {const response = await sendRequest<{ parameters: Record<string, any> }>('get_node_parameters', {node_id: memoryEdge.source});- const sessionId = response?.parameters?.sessionId || 'default';- setConnectedMemorySessionId(sessionId);+ const configuredSession = response?.parameters?.sessionId || '';+ // Auto-derive: use agent's nodeId if sessionId is empty or 'default'+ const actualSessionId = configuredSession && configuredSession !== 'default'+ ? configuredSession+ : nodeId; // nodeId is the agent's ID+ setConnectedMemorySessionId(actualSessionId);// Fetch compaction stats for this sessionsetCompactionLoading(true);@@ -205,11 +235,11 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({total: number;threshold: number;count: number;- }>('get_compaction_stats', { session_id: sessionId });+ }>('get_compaction_stats', { session_id: actualSessionId });if (statsResponse) {setCompactionStats({- session_id: statsResponse.session_id || sessionId,+ session_id: statsResponse.session_id || actualSessionId,total: statsResponse.total || 0,threshold: statsResponse.threshold || 100000,count: statsResponse.count || 0@@ -225,6 +255,28 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({fetchMemorySessionId();}, [nodeId, isAgentWithSkills, currentWorkflow, sendRequest]);+ // For Memory nodes: find which AI Agent this memory is connected TO+ // Used to display the auto-derived session ID+ useEffect(() => {+ if (!isMemoryNode || !currentWorkflow) {+ setConnectedAgentId(null);+ return;+ }++ const edges = currentWorkflow.edges || [];++ // Find edge where this memory node is the SOURCE and target handle is 'input-memory'+ const agentEdge = edges.find((edge: Edge) =>+ edge.source === nodeId && edge.targetHandle === 'input-memory'+ );++ if (agentEdge) {+ setConnectedAgentId(agentEdge.target);+ } else {+ setConnectedAgentId(null);+ }+ }, [nodeId, isMemoryNode, currentWorkflow]);+// Get connected skills for agent nodesuseEffect(() => {if (!isAgentWithSkills || !currentWorkflow) {@@ -388,6 +440,7 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({onConfigChange={(config) => onParameterChange('skillsConfig', config)}skillFolder={parameters.skillFolder || 'assistant'}onSkillFolderChange={(folder) => onParameterChange('skillFolder', folder)}+ nodeId={nodeId}/></div>) : (@@ -441,6 +494,7 @@ const MiddleSection: React.FC<MiddleSectionProps> = ({onParameterChange={onParameterChange}onClosePanel={() => {}}isLoadingParameters={isLoadingParameters}+ connectedAgentId={isMemoryNode ? connectedAgentId : undefined}/></div>);diff --git a/client/src/components/ui/QRCodeDisplay.tsx b/client/src/components/ui/QRCodeDisplay.tsxindex 5725b9b5..c245f4c1 100644--- a/client/src/components/ui/QRCodeDisplay.tsx+++ b/client/src/components/ui/QRCodeDisplay.tsx@@ -49,7 +49,7 @@ const isBase64Image = (value: string): boolean => {const QRCodeDisplay: React.FC<QRCodeDisplayProps> = ({value,isConnected = false,- size = 160,+ size = 280,loading = false,connectedTitle = 'Connected',connectedSubtitle = 'No QR code needed',diff --git a/client/src/components/ui/SettingsPanel.tsx b/client/src/components/ui/SettingsPanel.tsxindex 65b17fa1..ee58ad9b 100644--- a/client/src/components/ui/SettingsPanel.tsx+++ b/client/src/components/ui/SettingsPanel.tsx@@ -10,6 +10,7 @@ export interface WorkflowSettings {sidebarDefaultOpen: boolean;componentPaletteDefaultOpen: boolean;consolePanelDefaultOpen: boolean;+ memoryWindowSize: number;}export const defaultSettings: WorkflowSettings = {@@ -18,6 +19,7 @@ export const defaultSettings: WorkflowSettings = {sidebarDefaultOpen: true,componentPaletteDefaultOpen: true,consolePanelDefaultOpen: false,+ memoryWindowSize: 100,};interface SettingsPanelProps {@@ -57,6 +59,7 @@ const SettingsPanel: React.FC<SettingsPanelProps> = ({sidebarDefaultOpen: dbSettings.sidebar_default_open ?? defaultSettings.sidebarDefaultOpen,componentPaletteDefaultOpen: dbSettings.component_palette_default_open ?? defaultSettings.componentPaletteDefaultOpen,consolePanelDefaultOpen: dbSettings.console_panel_default_open ?? defaultSettings.consolePanelDefaultOpen,+ memoryWindowSize: dbSettings.memory_window_size ?? defaultSettings.memoryWindowSize,});}} catch (error) {@@ -76,6 +79,7 @@ const SettingsPanel: React.FC<SettingsPanelProps> = ({sidebar_default_open: newSettings.sidebarDefaultOpen,component_palette_default_open: newSettings.componentPaletteDefaultOpen,console_panel_default_open: newSettings.consolePanelDefaultOpen,+ memory_window_size: newSettings.memoryWindowSize,}});if (showMessage) {@@ -391,6 +395,31 @@ const SettingsPanel: React.FC<SettingsPanelProps> = ({</div>)}</div>++ {/* Memory Settings */}+ <div style={sectionStyle}>+ <div style={sectionHeaderStyle}>+ <div style={{ ...sectionIconStyle, backgroundColor: `${theme.dracula.purple}20` }}>{'\uD83E\uDDE0'}</div>+ <div style={sectionTitleStyle}>Memory</div>+ </div>++ <div style={settingRowStyle}>+ <div style={settingLabelStyle}>+ <div style={labelTextStyle}>Default Window Size</div>+ <div style={descriptionStyle}>Number of message pairs to keep in short-term memory (1-100)</div>+ </div>+ <InputNumber+ size="small"+ min={1}+ max={100}+ step={1}+ value={settings.memoryWindowSize}+ onChange={(value) => handleChange('memoryWindowSize', value ?? 100)}+ disabled={isSaving}+ style={{ width: 80 }}+ />+ </div>+ </div></div></div>diff --git a/client/src/contexts/WebSocketContext.tsx b/client/src/contexts/WebSocketContext.tsxindex ddc84d39..9590ad3e 100644--- a/client/src/contexts/WebSocketContext.tsx+++ b/client/src/contexts/WebSocketContext.tsx@@ -201,6 +201,7 @@ interface WebSocketContextValue {// Status dataandroidStatus: AndroidStatus;+ setAndroidStatus: React.Dispatch<React.SetStateAction<AndroidStatus>>;whatsappStatus: WhatsAppStatus;whatsappMessages: WhatsAppMessage[]; // History of received messageslastWhatsAppMessage: WhatsAppMessage | null; // Most recent message@@ -1956,6 +1957,7 @@ export const WebSocketProvider: React.FC<{ children: React.ReactNode }> = ({ chi// Status dataandroidStatus,+ setAndroidStatus,whatsappStatus,whatsappMessages,lastWhatsAppMessage,diff --git a/client/src/factories/baseChatModelFactory.ts b/client/src/factories/baseChatModelFactory.tsindex 04686a6b..225de43f 100644--- a/client/src/factories/baseChatModelFactory.ts+++ b/client/src/factories/baseChatModelFactory.ts@@ -152,8 +152,8 @@ export const STANDARD_PARAMETERS = {displayName: 'Maximum Tokens',name: 'maxTokens',type: 'number' as const,- default: 1000,- typeOptions: { minValue: 1, maxValue: 8192 },+ default: 4096,+ typeOptions: { minValue: 1, maxValue: 128000 },description: 'Maximum number of tokens to generate'},diff --git a/client/src/hooks/useApiKeys.ts b/client/src/hooks/useApiKeys.tsindex 897cfa36..979cf16b 100644--- a/client/src/hooks/useApiKeys.ts+++ b/client/src/hooks/useApiKeys.ts@@ -14,6 +14,15 @@ export interface ApiKeyValidationResult {models?: string[];}+export interface ProviderDefaults {+ temperature: number;+ max_tokens: number;+ thinking_enabled: boolean;+ thinking_budget: number;+ reasoning_effort: 'low' | 'medium' | 'high';+ reasoning_format: 'parsed' | 'hidden';+}+export interface UseApiKeysResult {// Validate and store API keyvalidateApiKey: (provider: string, apiKey: string) => Promise<ApiKeyValidationResult>;@@ -39,6 +48,10 @@ export interface UseApiKeysResult {// Get AI models for a providergetAiModels: (provider: string, apiKey: string) => Promise<string[]>;+ // Provider defaults+ getProviderDefaults: (provider: string) => Promise<ProviderDefaults>;+ saveProviderDefaults: (provider: string, defaults: ProviderDefaults) => Promise<boolean>;+// StateisValidating: boolean;validationError: string | null;@@ -53,6 +66,7 @@ export const useApiKeys = (): UseApiKeysResult => {deleteApiKey: wsDeleteApiKey,validateMapsKey: wsValidateMapsKey,getAiModels: wsGetAiModels,+ sendRequest,isConnected} = useWebSocket();@@ -222,6 +236,50 @@ export const useApiKeys = (): UseApiKeysResult => {}}, [wsGetAiModels]);+ /**+ * Get default parameters for a provider+ */+ const getProviderDefaults = useCallback(async (+ provider: string+ ): Promise<ProviderDefaults> => {+ const defaultValues: ProviderDefaults = {+ temperature: 0.7,+ max_tokens: 4096,+ thinking_enabled: false,+ thinking_budget: 2048,+ reasoning_effort: 'medium',+ reasoning_format: 'parsed',+ };++ if (!isConnected) return defaultValues;++ try {+ const response = await sendRequest<{ defaults: ProviderDefaults }>('get_provider_defaults', { provider });+ return response?.defaults || defaultValues;+ } catch (error) {+ console.warn(`Error fetching provider defaults for ${provider}:`, error);+ return defaultValues;+ }+ }, [sendRequest, isConnected]);++ /**+ * Save default parameters for a provider+ */+ const saveProviderDefaults = useCallback(async (+ provider: string,+ defaults: ProviderDefaults+ ): Promise<boolean> => {+ if (!isConnected) return false;++ try {+ const response = await sendRequest<{ success: boolean }>('save_provider_defaults', { provider, defaults });+ return response?.success || false;+ } catch (error) {+ console.warn(`Error saving provider defaults for ${provider}:`, error);+ return false;+ }+ }, [sendRequest, isConnected]);+return {validateApiKey,saveApiKey,@@ -231,6 +289,8 @@ export const useApiKeys = (): UseApiKeysResult => {removeApiKey,validateGoogleMapsKey,getAiModels,+ getProviderDefaults,+ saveProviderDefaults,isValidating,validationError,isConnecteddiff --git a/client/src/nodeDefinitions/aiAgentNodes.ts b/client/src/nodeDefinitions/aiAgentNodes.tsindex 63a0dfca..00fa923a 100644--- a/client/src/nodeDefinitions/aiAgentNodes.ts+++ b/client/src/nodeDefinitions/aiAgentNodes.ts@@ -74,6 +74,18 @@ export const aiAgentNodes: Record<string, INodeTypeDescription> = {{name: 'Google (Gemini)',value: 'gemini'+ },+ {+ name: 'Groq',+ value: 'groq'+ },+ {+ name: 'OpenRouter',+ value: 'openrouter'+ },+ {+ name: 'Cerebras',+ value: 'cerebras'}],default: 'openai',@@ -133,8 +145,8 @@ export const aiAgentNodes: Record<string, INodeTypeDescription> = {displayName: 'Maximum Tokens',name: 'maxTokens',type: 'number',- default: 1000,- typeOptions: { minValue: 1, maxValue: 8192 },+ default: 4096,+ typeOptions: { minValue: 1, maxValue: 128000 },description: 'Maximum number of tokens to generate'},{@@ -244,6 +256,10 @@ export const aiAgentNodes: Record<string, INodeTypeDescription> = {{name: 'OpenRouter',value: 'openrouter'+ },+ {+ name: 'Cerebras',+ value: 'cerebras'}],default: 'openai',@@ -298,8 +314,8 @@ export const aiAgentNodes: Record<string, INodeTypeDescription> = {displayName: 'Maximum Tokens',name: 'maxTokens',type: 'number',- default: 1000,- typeOptions: { minValue: 1, maxValue: 8192 },+ default: 4096,+ typeOptions: { minValue: 1, maxValue: 128000 },description: 'Maximum number of tokens to generate'},{@@ -342,7 +358,7 @@ export const aiAgentNodes: Record<string, INodeTypeDescription> = {displayName: 'Simple Memory',name: 'simpleMemory',icon: '🧠',- group: ['skill', 'memory'], // 'skill' = appears in AI Skills category+ group: ['tool', 'memory'], // 'tool' = appears in AI Tools categoryversion: 1,description: 'Markdown-based conversation memory with optional vector DB for long-term retrieval',defaults: { name: 'Memory', color: '#8b5cf6' },@@ -355,20 +371,21 @@ export const aiAgentNodes: Record<string, INodeTypeDescription> = {}],properties: [{- displayName: 'Session ID',+ displayName: 'Session ID (Override)',name: 'sessionId',type: 'string',- default: 'default',- required: true,- description: 'Unique identifier for conversation session'+ default: '',+ required: false,+ description: 'Leave empty to auto-use connected agent ID. Set manually to share memory across agents.',+ placeholder: 'Auto (uses agent ID)'},{displayName: 'Window Size',name: 'windowSize',type: 'number',- default: 10,+ default: 100,typeOptions: { minValue: 1, maxValue: 100 },- description: 'Number of message pairs to keep in short-term memory'+ description: 'Number of message pairs to keep in short-term memory (uses global default from Settings)'},{displayName: 'Conversation History',diff --git a/client/src/nodeDefinitions/skillNodes.ts b/client/src/nodeDefinitions/skillNodes.tsindex facd8394..2f43619a 100644--- a/client/src/nodeDefinitions/skillNodes.ts+++ b/client/src/nodeDefinitions/skillNodes.ts@@ -3,184 +3,23 @@ import {INodeTypeDescription,NodeConnectionType} from '../types/INodeProperties';-import { WHATSAPP_CONNECT_ICON } from './whatsappNodes';// ============================================================================// SKILL NODES - Connect to Zeenie's input-skill handle// ============================================================================--// Factory function for creating skill node definitions-export function createSkillNode(config: {- name: string;- displayName: string;- icon: string;- color: string;- skillName: string;- description: string;- subtitle: string;- properties?: any[];-}): INodeTypeDescription {- return {- displayName: config.displayName,- name: config.name,- icon: config.icon,- group: ['skill'],- version: 1,- subtitle: config.subtitle,- description: config.description,- defaults: { name: config.displayName, color: config.color },- inputs: [], // No input - skill node is passive- outputs: [{- name: 'skill',- displayName: 'Skill',- type: 'main' as NodeConnectionType,- description: 'Connect to Zeenie skill handle'- }],- properties: [- {- displayName: 'Skill Name',- name: 'skillName',- type: 'string',- default: config.skillName,- required: true,- description: 'Internal skill identifier',- typeOptions: { readonly: true }- },- {- displayName: 'Instructions',- name: 'instructions',- type: 'string',- default: '',- description: 'Skill instructions in Markdown format. Loaded from SKILL.md file.',- typeOptions: {- rows: 20,- editor: 'code',- editorLanguage: 'markdown'- }- },- ...(config.properties || [])- ]- };-}+// Master Skill node aggregates built-in skills from server/skills/ folders+// and user-created skills from database. Supports inline skill creation/editing.export const skillNodes: Record<string, INodeTypeDescription> = {- // Assistant Personality Skill - Default assistant personality- assistantPersonality: createSkillNode({- name: 'assistantPersonality',- displayName: 'Assistant Personality',- icon: '✨',- color: '#D97706',- skillName: 'assistant-personality',- subtitle: 'AI Personality',- description: 'Default assistant personality with helpful, harmless, and honest responses'- }),-- // WhatsApp Skill - Send and receive messages- whatsappSkill: createSkillNode({- name: 'whatsappSkill',- displayName: 'WhatsApp Skill',- icon: WHATSAPP_CONNECT_ICON,- color: '#25D366',- skillName: 'whatsapp-skill',- subtitle: 'Messaging',- description: 'Send and receive WhatsApp messages via Zeenie'- }),-- // Memory Skill - Remember and recall information- memorySkill: createSkillNode({- name: 'memorySkill',- displayName: 'Memory Skill',- icon: '🧠',- color: '#8B5CF6',- skillName: 'memory-skill',- subtitle: 'Long-term Memory',- description: 'Remember information across conversations with short and long-term memory',- properties: [- {- displayName: 'Default Session',- name: 'defaultSession',- type: 'string',- default: 'default',- description: 'Default session ID for memory storage'- }- ]- }),-- // Maps Skill - Location and geocoding services- mapsSkill: createSkillNode({- name: 'mapsSkill',- displayName: 'Maps Skill',- icon: '🗺️',- color: '#4285F4',- skillName: 'maps-skill',- subtitle: 'Location Services',- description: 'Geocoding, nearby places, and map creation via Google Maps'- }),-- // HTTP Skill - Make API requests- httpSkill: createSkillNode({- name: 'httpSkill',- displayName: 'HTTP Skill',- icon: '🌐',- color: '#EF4444',- skillName: 'http-skill',- subtitle: 'API Requests',- description: 'Make HTTP requests to external APIs and web services'- }),-- // Scheduler Skill - Timers and cron jobs- schedulerSkill: createSkillNode({- name: 'schedulerSkill',- displayName: 'Scheduler Skill',- icon: '⏰',- color: '#10B981',- skillName: 'scheduler-skill',- subtitle: 'Task Scheduling',- description: 'Schedule tasks with timers and cron expressions'- }),-- // Android Skill - Device control- androidSkill: createSkillNode({- name: 'androidSkill',- displayName: 'Android Skill',- icon: '📱',- color: '#3DDC84',- skillName: 'android-skill',- subtitle: 'Device Control',- description: 'Control Android devices - battery, wifi, bluetooth, apps, location, camera'- }),-- // Code Skill - Execute code- codeSkill: createSkillNode({- name: 'codeSkill',- displayName: 'Code Skill',- icon: '💻',- color: '#F59E0B',- skillName: 'code-skill',- subtitle: 'Code Execution',- description: 'Execute Python or JavaScript code for calculations and data processing'- }),-- // Web Search Skill - Search the web for information- webSearchSkill: createSkillNode({- name: 'webSearchSkill',- displayName: 'Web Search Skill',- icon: '🔍',- color: '#bd93f9',- skillName: 'web-search-skill',- subtitle: 'Web Search',- description: 'Search the web for current information, news, facts, and real-time data'- }),-// Master Skill - Aggregates multiple skills with enable/disable togglesmasterSkill: {displayName: 'Master Skill',name: 'masterSkill',icon: '🎯',- group: ['skill'],+ group: ['tool'], // Appears in AI Tools categoryversion: 1,subtitle: 'Skill Aggregator',- description: 'Combine multiple skills with enable/disable toggles in a single connection',+ description: 'Combine built-in and custom skills with enable/disable toggles',defaults: { name: 'Master Skill', color: '#9333EA' },inputs: [],outputs: [{@@ -205,54 +44,12 @@ export const skillNodes: Record<string, INodeTypeDescription> = {description: 'Enabled skills and their instructions'}]- },-- // Custom Skill - User-created skill- customSkill: {- displayName: 'Custom Skill',- name: 'customSkill',- icon: '⭐',- group: ['skill'],- version: 1,- subtitle: 'User-defined',- description: 'User-created custom skill with configurable capabilities',- defaults: { name: 'Custom Skill', color: '#6366F1' },- inputs: [],- outputs: [{- name: 'skill',- displayName: 'Skill',- type: 'main' as NodeConnectionType,- description: 'Connect to Zeenie skill handle'- }],- properties: [- {- displayName: 'Skill',- name: 'selectedSkill',- type: 'options',- default: '',- required: true,- description: 'Select a user-created skill',- typeOptions: {- dynamicOptions: true- }- }- ]}};// List of skill node types for easy identificationexport const SKILL_NODE_TYPES = [- 'assistantPersonality',- 'whatsappSkill',- 'memorySkill',- 'mapsSkill',- 'httpSkill',- 'schedulerSkill',- 'androidSkill',- 'codeSkill',- 'webSearchSkill',- 'masterSkill',- 'customSkill'+ 'masterSkill'];// Export for nodeDefinitions.tsdiff --git a/client/src/nodeDefinitions/specializedAgentNodes.ts b/client/src/nodeDefinitions/specializedAgentNodes.tsindex afdbad41..6a79721f 100644--- a/client/src/nodeDefinitions/specializedAgentNodes.ts+++ b/client/src/nodeDefinitions/specializedAgentNodes.ts@@ -69,7 +69,8 @@ export const AI_AGENT_PROPERTIES: INodeProperties[] = [{ name: 'Anthropic (Claude)', value: 'anthropic' },{ name: 'Google (Gemini)', value: 'gemini' },{ name: 'Groq', value: 'groq' },- { name: 'OpenRouter', value: 'openrouter' }+ { name: 'OpenRouter', value: 'openrouter' },+ { name: 'Cerebras', value: 'cerebras' }],default: 'openai',description: 'The AI provider to use (configure API keys in Credentials)'@@ -124,8 +125,8 @@ export const AI_AGENT_PROPERTIES: INodeProperties[] = [displayName: 'Maximum Tokens',name: 'maxTokens',type: 'number',- default: 1000,- typeOptions: { minValue: 1, maxValue: 8192 },+ default: 4096,+ typeOptions: { minValue: 1, maxValue: 128000 },description: 'Maximum number of tokens to generate'},{@@ -314,6 +315,36 @@ export const specializedAgentNodes: Record<string, INodeTypeDescription> = {inputs: AI_AGENT_INPUTS,outputs: AI_AGENT_OUTPUTS,properties: AI_AGENT_PROPERTIES+ },++ // Autonomous Agent - AI Agent for autonomous operations with Code Mode patterns+ autonomous_agent: {+ displayName: 'Autonomous Agent',+ name: 'autonomous_agent',+ icon: '🎯',+ group: ['agent', 'ai'],+ version: 1,+ subtitle: 'Autonomous Operations',+ description: 'AI Agent specialized for autonomous operations. Uses Code Mode patterns (81-98% token savings), agentic loops, progressive discovery, error recovery, and multi-tool orchestration.',+ defaults: { name: 'Autonomous Agent', color: dracula.purple },+ inputs: AI_AGENT_INPUTS,+ outputs: AI_AGENT_OUTPUTS,+ properties: AI_AGENT_PROPERTIES+ },++ // Orchestrator Agent - Coordinates multiple agents for complex workflows+ orchestrator_agent: {+ displayName: 'Orchestrator Agent',+ name: 'orchestrator_agent',+ icon: '🎼',+ group: ['agent', 'ai'],+ version: 1,+ subtitle: 'Agent Coordination',+ description: 'AI Agent specialized for orchestrating multiple agents. Coordinates complex multi-agent workflows by delegating tasks to specialized agents and synthesizing their results.',+ defaults: { name: 'Orchestrator Agent', color: dracula.cyan },+ inputs: AI_AGENT_INPUTS,+ outputs: AI_AGENT_OUTPUTS,+ properties: AI_AGENT_PROPERTIES}};@@ -322,4 +353,4 @@ export const specializedAgentNodes: Record<string, INodeTypeDescription> = {// ============================================================================// List of specialized agent node types for identification-export const SPECIALIZED_AGENT_TYPES = ['android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent'];+export const SPECIALIZED_AGENT_TYPES = ['android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent', 'autonomous_agent', 'orchestrator_agent'];diff --git a/docs-internal/specialized_agent_node_creation.md b/docs-internal/specialized_agent_node_creation.mdindex 5585b1bb..eae028db 100644--- a/docs-internal/specialized_agent_node_creation.md+++ b/docs-internal/specialized_agent_node_creation.md@@ -58,6 +58,8 @@ All agent colors use centralized dracula theme constants from `client/src/styles| `productivity_agent` | Productivity Agent | Productivity workflows agent | `dracula.cyan` || `payments_agent` | Payments Agent | Payment processing agent | `dracula.green` || `consumer_agent` | Consumer Agent | Consumer interactions agent | `dracula.purple` |+| `autonomous_agent` | Autonomous Agent | Autonomous operations with Code Mode patterns | `dracula.purple` |+| `orchestrator_agent` | Orchestrator Agent | Multi-agent coordination and task delegation | `dracula.cyan` |## Shared AI Agent Properties@@ -754,13 +756,13 @@ Memory connected to `input-memory` provides conversation history via a two-tier### Shared Handler Routing-All 10 specialized agents route to the same backend handler:+All 12 specialized agents route to the same backend handler:```python# In node_executor.py'android_agent': partial(handle_chat_agent, ai_service=..., database=...),'coding_agent': partial(handle_chat_agent, ai_service=..., database=...),-# ... all 10 specialized agents use handle_chat_agent+# ... all 12 specialized agents use handle_chat_agent```This means specialization is purely UI/configuration, not backend behavioral. All agents share:diff --git a/package-lock.json b/package-lock.jsonindex 5b81a8ae..41f0a48d 100644--- a/package-lock.json+++ b/package-lock.json@@ -27,7 +27,7 @@},"client": {"name": "react-flow-client",- "version": "0.0.12",+ "version": "0.0.13","dependencies": {"@ant-design/icons": "^6.0.2","@eslint/js": "^9.33.0",diff --git a/package.json b/package.jsonindex 8a59c3db..3e7747fa 100644--- a/package.json+++ b/package.json@@ -70,6 +70,7 @@"whatsapp:stop": "whatsapp-rpc stop","whatsapp:status": "whatsapp-rpc status","whatsapp:api": "whatsapp-rpc api --foreground",+ "whatsapp:build": "whatsapp-rpc build","build": "node scripts/build.js","clean": "node scripts/clean.js","stop": "node scripts/stop.js",diff --git a/scripts/build.js b/scripts/build.jsindex b6e42869..adb04c5b 100644--- a/scripts/build.js+++ b/scripts/build.js@@ -84,7 +84,11 @@ function installPython() {process.exit(1);}} else {- if (runSilent('apt --version')) {+ // Linux package managers+ if (runSilent('apk --version')) {+ // Alpine Linux+ run('apk add --no-cache python3 py3-pip');+ } else if (runSilent('apt --version')) {run('sudo apt update && sudo apt install -y python3.12 python3.12-venv');} else if (runSilent('dnf --version')) {run('sudo dnf install -y python3.12');@@ -97,17 +101,25 @@ function installPython() {}}-function installUv() {+function installUv(pythonCmd) {console.log(' Installing uv...');if (isWindows) {- run('powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"');- } else {- run('curl -LsSf https://astral.sh/uv/install.sh | sh');- }- if (isWindows) {- process.env.PATH = `${process.env.USERPROFILE}\\.local\\bin;${process.env.PATH}`;+ // Windows: use pip (simple and works)+ run(`${pythonCmd} -m pip install uv`);+ } else if (isMac) {+ // macOS: use pip (no PEP 668 issues with Homebrew Python)+ run(`${pythonCmd} -m pip install uv`);+ } else if (runSilent('apk --version')) {+ // Alpine: use pip with --break-system-packages+ run(`${pythonCmd} -m pip install uv --break-system-packages`);} else {- process.env.PATH = `${process.env.HOME}/.local/bin:${process.env.PATH}`;+ // Other Linux: try pip first, fall back to curl installer+ if (runSilent(`${pythonCmd} -m pip install uv --break-system-packages`)) {+ // pip worked+ } else {+ run('curl -LsSf https://astral.sh/uv/install.sh | sh');+ process.env.PATH = `${process.env.HOME}/.local/bin:${process.env.PATH}`;+ }}}@@ -156,7 +168,7 @@ let uvVersion = getVersion('uv --version');if (uvVersion) {console.log(` uv: ${uvVersion}`);} else {- installUv();+ installUv(pyCmd);uvVersion = getVersion('uv --version');if (uvVersion) {console.log(` uv: ${uvVersion}`);@@ -210,36 +222,15 @@ try {// Step 4: Install Python dependenciesconsole.log('[4/5] Installing Python dependencies...');const serverDir = resolve(ROOT, 'server');- run('uv venv', serverDir);+ // Check if .venv exists, skip creation if it does+ if (!existsSync(resolve(serverDir, '.venv'))) {+ run('uv venv', serverDir);+ }run('uv sync', serverDir);- // Step 5: Ensure WhatsApp RPC binary is downloaded- // The whatsapp-rpc postinstall skips download in CI, so we run it explicitly.- const whatsappBin = resolve(ROOT, 'node_modules', 'whatsapp-rpc', 'bin',- isWindows ? 'whatsapp-rpc-server.exe' : 'whatsapp-rpc-server');- const whatsappDownloadScript = resolve(ROOT, 'node_modules', 'whatsapp-rpc', 'scripts', 'download-binary.js');- if (!existsSync(whatsappBin) && existsSync(whatsappDownloadScript)) {- console.log('[5/5] Downloading WhatsApp RPC binary...');- try {- // Run download script without CI env vars so it doesn't skip- const dlEnv = { ...process.env };- delete dlEnv.CI;- delete dlEnv.GITHUB_ACTIONS;- const scriptPath = whatsappDownloadScript.replace(/\\/g, '/');- execSync(`node "${scriptPath}"`, {- cwd: resolve(ROOT, 'node_modules', 'whatsapp-rpc'),- stdio: 'inherit',- shell: true,- env: dlEnv- });- } catch (e) {- console.log('Warning: Failed to download WhatsApp RPC binary. WhatsApp features will be unavailable.');- }- } else if (existsSync(whatsappBin)) {- console.log('[5/5] WhatsApp RPC binary already present');- } else {- console.log('[5/5] WhatsApp RPC package not found, skipping binary download');- }+ // Step 5: Verify WhatsApp RPC package+ console.log('[5/5] Verifying WhatsApp RPC...');+ run('whatsapp-rpc status', ROOT);console.log('\nBuild complete.');diff --git a/scripts/install.js b/scripts/install.jsindex 7831f6e6..6102b78c 100644--- a/scripts/install.js+++ b/scripts/install.js@@ -120,6 +120,9 @@ function installUv(pythonCmd) {} else if (isMac) {// Use pip on macOS (no PEP 668 issues with Homebrew Python)run(`${pythonCmd} -m pip install uv`);+ } else if (runSilent('apk --version')) {+ // Alpine: use pip (apk doesn't have uv package)+ run(`${pythonCmd} -m pip install uv --break-system-packages`);} else {// Linux: use official installer (avoids PEP 668 externally-managed-environment)run('curl -LsSf https://astral.sh/uv/install.sh | sh');@@ -219,7 +222,11 @@ try {// Install Python dependencies (always needed - venv not included in package)step++;console.log(`[${step}/${totalSteps}] Installing Python dependencies...`);- run('uv venv', serverDir); // 5 min default+ // Check if .venv exists, skip creation if it does+ const venvPath = resolve(serverDir, '.venv');+ if (!existsSync(venvPath)) {+ run('uv venv', serverDir); // 5 min default+ }run('uv sync', serverDir, 600000); // 10 min timeout// WhatsApp RPC is now an npm dependency - binary downloaded via postinstalldiff --git a/scripts/start.js b/scripts/start.jsindex 67da5dfd..79583447 100644--- a/scripts/start.js+++ b/scripts/start.js@@ -227,14 +227,8 @@ if (isProduction) {// Python backend: use gunicorn in daemon mode, uvicorn in developmentservices.push(isDaemonMode ? 'npm:python:daemon' : 'npm:python:start');-// WhatsApp RPC: only start if the pre-built binary exists-const whatsappBin = resolve(ROOT, 'node_modules', 'whatsapp-rpc', 'bin',- isWindows ? 'whatsapp-rpc-server.exe' : 'whatsapp-rpc-server');-if (existsSync(whatsappBin)) {- services.push('npm:whatsapp:api');-} else {- log('WhatsApp RPC binary not found, skipping. Run "npm run build" to download it.');-}+// WhatsApp RPC service - npm handles package availability+services.push('npm:whatsapp:api');if (config.temporalEnabled) {services.push('npm:temporal:worker');diff --git a/server/constants.py b/server/constants.pyindex d09c2c23..d2380bf5 100644--- a/server/constants.py+++ b/server/constants.py@@ -32,6 +32,8 @@ AI_AGENT_TYPES: FrozenSet[str] = frozenset(['productivity_agent','payments_agent','consumer_agent',+ 'autonomous_agent',+ 'orchestrator_agent',])AI_MEMORY_TYPES: FrozenSet[str] = frozenset([@@ -48,16 +50,9 @@ AI_TOOL_TYPES: FrozenSet[str] = frozenset([# Skill node types (connect to Zeenie's input-skill handle)# Skills provide SKILL.md context/instructions, not executed as workflow nodes+# masterSkill: Aggregates built-in skills and user-created skills with inline editingSKILL_NODE_TYPES: FrozenSet[str] = frozenset([- 'assistantPersonality',- 'whatsappSkill',- 'memorySkill',- 'mapsSkill',- 'httpSkill',- 'schedulerSkill',- 'androidSkill',- 'codeSkill',- 'customSkill',+ 'masterSkill',])# Toolkit node types (aggregate sub-nodes, n8n Sub-Node pattern)diff --git a/server/core/container.py b/server/core/container.pyindex 84d2ddc2..c173c33d 100644--- a/server/core/container.py+++ b/server/core/container.py@@ -13,9 +13,23 @@ from services.auth import AuthServicefrom services.text import TextServicefrom services.android_service import AndroidServicefrom services.user_auth import UserAuthService-from services.temporal import TemporalClientWrapperfrom services.compaction import init_compaction_service+# Temporal is optional - import with fallback+try:+ from services.temporal import TemporalClientWrapper+ _TEMPORAL_AVAILABLE = True+except ImportError:+ TemporalClientWrapper = None+ _TEMPORAL_AVAILABLE = False+++def _create_temporal_client(server_address: str, namespace: str):+ """Factory function for temporal client - returns None if not available."""+ if not _TEMPORAL_AVAILABLE or TemporalClientWrapper is None:+ return None+ return TemporalClientWrapper(server_address=server_address, namespace=namespace)+class Container(containers.DeclarativeContainer):"""Application dependency injection container."""@@ -41,9 +55,9 @@ class Container(containers.DeclarativeContainer):database=database)- # Temporal client (lazy - only created when temporal_enabled)+ # Temporal client (lazy - returns None if temporalio not installed)temporal_client = providers.Singleton(- TemporalClientWrapper,+ _create_temporal_client,server_address=settings.provided.temporal_server_address,namespace=settings.provided.temporal_namespace,)diff --git a/server/core/database.py b/server/core/database.pyindex e7b4949a..fb6d8d98 100644--- a/server/core/database.py+++ b/server/core/database.py@@ -12,7 +12,7 @@ from core.config import Settingsfrom models.database import (NodeParameter, Workflow, Execution, APIKey, APIKeyValidation, NodeOutput,ConversationMessage, ToolSchema, UserSkill, ChatMessage, UserSettings,- TokenUsageMetric, CompactionEvent, SessionTokenState+ TokenUsageMetric, CompactionEvent, SessionTokenState, ProviderDefaults)from models.cache import CacheEntry # SQLite-backed cache for Redis alternativefrom models.auth import User # Import User model to ensure table creation@@ -1482,6 +1482,70 @@ class Database:logger.error("Failed to save user settings", user_id=user_id, error=str(e))return False+ # ============================================================================+ # Provider Defaults+ # ============================================================================++ async def get_provider_defaults(self, provider: str) -> Optional[Dict[str, Any]]:+ """Get default parameters for a provider."""+ try:+ async with self.get_session() as session:+ stmt = select(ProviderDefaults).where(ProviderDefaults.provider == provider.lower())+ result = await session.execute(stmt)+ record = result.scalar_one_or_none()+ if not record:+ return None+ return {+ "temperature": record.temperature,+ "max_tokens": record.max_tokens,+ "thinking_enabled": record.thinking_enabled,+ "thinking_budget": record.thinking_budget,+ "reasoning_effort": record.reasoning_effort,+ "reasoning_format": record.reasoning_format,+ }+ except Exception as e:+ logger.error("Failed to get provider defaults", provider=provider, error=str(e))+ return None++ async def save_provider_defaults(self, provider: str, defaults: Dict[str, Any]) -> bool:+ """Upsert provider defaults."""+ try:+ async with self.get_session() as session:+ stmt = select(ProviderDefaults).where(ProviderDefaults.provider == provider.lower())+ result = await session.execute(stmt)+ record = result.scalar_one_or_none()++ now = datetime.now(timezone.utc)+ if record:+ record.temperature = defaults.get("temperature", record.temperature)+ record.max_tokens = defaults.get("max_tokens", record.max_tokens)+ record.thinking_enabled = defaults.get("thinking_enabled", record.thinking_enabled)+ record.thinking_budget = defaults.get("thinking_budget", record.thinking_budget)+ record.reasoning_effort = defaults.get("reasoning_effort", record.reasoning_effort)+ record.reasoning_format = defaults.get("reasoning_format", record.reasoning_format)+ record.updated_at = now+ else:+ record = ProviderDefaults(+ provider=provider.lower(),+ temperature=defaults.get("temperature", 0.7),+ max_tokens=defaults.get("max_tokens", 1000),+ thinking_enabled=defaults.get("thinking_enabled", False),+ thinking_budget=defaults.get("thinking_budget", 2048),+ reasoning_effort=defaults.get("reasoning_effort", "medium"),+ reasoning_format=defaults.get("reasoning_format", "parsed"),+ created_at=now,+ updated_at=now+ )+ session.add(record)++ await session.commit()+ logger.info(f"[DB] Provider defaults saved for: {provider}")+ return True++ except Exception as e:+ logger.error("Failed to save provider defaults", provider=provider, error=str(e))+ return False+# ============================================================================# Token Usage Metrics# ============================================================================@@ -1557,7 +1621,10 @@ class Database:async def get_or_create_session_token_state(self, session_id: str) -> Dict[str, Any]:- """Get or create token state for a session."""+ """Get or create token state for a session.++ Handles race conditions when multiple requests try to create the same session.+ """try:async with self.get_session() as session:stmt = select(SessionTokenState).where(@@ -1567,13 +1634,34 @@ class Database:state = result.scalar_one_or_none()if not state:- state = SessionTokenState(- session_id=session_id,- updated_at=datetime.now(timezone.utc)- )- session.add(state)- await session.commit()- await session.refresh(state)+ try:+ state = SessionTokenState(+ session_id=session_id,+ updated_at=datetime.now(timezone.utc)+ )+ session.add(state)+ await session.commit()+ await session.refresh(state)+ except IntegrityError:+ # Race condition: another request created the session+ # Rollback and fetch the existing record+ await session.rollback()+ result = await session.execute(stmt)+ state = result.scalar_one_or_none()+ if not state:+ # Shouldn't happen, but return defaults if it does+ return {+ "session_id": session_id,+ "cumulative_input_tokens": 0,+ "cumulative_output_tokens": 0,+ "cumulative_cache_tokens": 0,+ "cumulative_reasoning_tokens": 0,+ "cumulative_total": 0,+ "last_compaction_at": None,+ "compaction_count": 0,+ "custom_threshold": None,+ "compaction_enabled": True+ }return {"session_id": state.session_id,diff --git a/server/main.py b/server/main.pyindex 11d012af..e36664ca 100644--- a/server/main.py+++ b/server/main.py@@ -132,6 +132,10 @@ async def lifespan(app: FastAPI):from services.temporal import TemporalClientWrapper, TemporalExecutorfrom services.temporal.worker import TemporalWorkerManager+ # Check if temporal is actually available (temporalio installed)+ if TemporalClientWrapper is None:+ raise ImportError("temporalio not installed")+logger.info("Initializing Temporal integration",server_address=settings.temporal_server_address,@@ -141,6 +145,8 @@ async def lifespan(app: FastAPI):# Connect Temporal clienttemporal_client_wrapper = container.temporal_client()+ if temporal_client_wrapper is None:+ raise ImportError("temporalio not installed")temporal_client = await temporal_client_wrapper.connect()# Create and set the Temporal executor on WorkflowService@@ -178,7 +184,9 @@ async def lifespan(app: FastAPI):# Disconnect Temporal client if connectedif settings.temporal_enabled:try:- await container.temporal_client().disconnect()+ temporal_client_wrapper = container.temporal_client()+ if temporal_client_wrapper is not None:+ await temporal_client_wrapper.disconnect()except Exception:pass@@ -285,9 +293,11 @@ async def health_check():}if settings.temporal_enabled:try:- temporal_status["connected"] = container.temporal_client().is_connected- temporal_status["server_address"] = settings.temporal_server_address- temporal_status["task_queue"] = settings.temporal_task_queue+ temporal_client_wrapper = container.temporal_client()+ if temporal_client_wrapper is not None:+ temporal_status["connected"] = temporal_client_wrapper.is_connected+ temporal_status["server_address"] = settings.temporal_server_address+ temporal_status["task_queue"] = settings.temporal_task_queueexcept Exception:passdiff --git a/server/models/database.py b/server/models/database.pyindex 57029658..b31850f9 100644--- a/server/models/database.py+++ b/server/models/database.py@@ -263,6 +263,33 @@ class UserSettings(SQLModel, table=True):)+class ProviderDefaults(SQLModel, table=True):+ """LLM provider default parameters - persisted across server restarts.++ Database is the source of truth for provider-specific default parameters.+ These defaults are applied to new AI nodes and can be overridden per-node.+ """++ __tablename__ = "provider_defaults"++ id: Optional[int] = Field(default=None, primary_key=True)+ provider: str = Field(unique=True, index=True, max_length=50) # openai, anthropic, gemini, etc.+ temperature: float = Field(default=0.7)+ max_tokens: int = Field(default=1000)+ thinking_enabled: bool = Field(default=False)+ thinking_budget: int = Field(default=2048)+ reasoning_effort: str = Field(default="medium", max_length=20) # low, medium, high+ reasoning_format: str = Field(default="parsed", max_length=20) # parsed, hidden+ created_at: datetime = Field(+ default_factory=lambda: datetime.now(timezone.utc),+ sa_column=Column(DateTime(timezone=True), server_default=func.now())+ )+ updated_at: datetime = Field(+ default_factory=lambda: datetime.now(timezone.utc),+ sa_column=Column(DateTime(timezone=True), onupdate=func.now())+ )++# =============================================================================# Token Tracking and Memory Compaction# =============================================================================diff --git a/server/pyproject.toml b/server/pyproject.tomlindex 0e46aa52..9b76f51d 100644--- a/server/pyproject.toml+++ b/server/pyproject.toml@@ -45,9 +45,6 @@ dependencies = [# Dependency Injection"dependency-injector>=4.40.0",- # Temporal (distributed execution)- "temporalio>=1.7.0",-# Utils"python-dotenv>=1.0.0","apscheduler>=3.10.0",@@ -63,6 +60,9 @@ dev = ["pytest>=8.0.0","ruff>=0.8.0",]+temporal = [+ "temporalio>=1.7.0",+]extra-ai = ["langchain-cerebras>=0.1.0",]diff --git a/server/routers/websocket.py b/server/routers/websocket.pyindex 4fbade3b..44992416 100644--- a/server/routers/websocket.py+++ b/server/routers/websocket.py@@ -1146,6 +1146,7 @@ async def handle_android_relay_connect(data: Dict[str, Any], websocket: WebSocketry:client, error = await get_relay_client(url, api_key)if client:+ logger.info(f"[WebSocket] Android relay connect success, qr_data present: {bool(client.qr_data)}, session_token: {client.session_token}")return {"success": True,"connected": True,@@ -1636,8 +1637,8 @@ async def handle_get_skill_content(data: Dict[str, Any], websocket: WebSocket) -skill_name = data["skill_name"]skill_loader = get_skill_loader()- # Try to load the skill- skill = skill_loader.load_skill(skill_name)+ # Use load_skill_async which checks both filesystem and database+ skill = await skill_loader.load_skill_async(skill_name)if skill:return {"success": True,@@ -1649,20 +1650,6 @@ async def handle_get_skill_content(data: Dict[str, Any], websocket: WebSocket) -"timestamp": time.time()}- # Try loading from database for user skills- database = container.database()- user_skill = await database.get_user_skill(skill_name)- if user_skill:- return {- "success": True,- "skill_name": skill_name,- "instructions": user_skill.instructions,- "description": user_skill.description,- "allowed_tools": user_skill.allowed_tools.split(',') if user_skill.allowed_tools else [],- "is_builtin": False,- "timestamp": time.time()- }-return {"success": False, "error": f"Skill '{skill_name}' not found"}@@ -1894,7 +1881,8 @@ async def handle_delete_user_skill(data: Dict[str, Any], websocket: WebSocket) -"name": data["name"],"timestamp": time.time()})- return {"deleted": True, "name": data["name"], "timestamp": time.time()}+ logger.info(f"[Skills] Deleted user skill: {data['name']}")+ return {"success": True, "deleted": True, "name": data["name"], "timestamp": time.time()}return {"success": False, "error": f"Skill '{data['name']}' not found"}@@ -2011,6 +1999,41 @@ async def handle_save_user_settings(data: Dict[str, Any], websocket: WebSocket)return {"success": False, "error": "Failed to save settings"}+# ============================================================================+# Provider Defaults Handlers+# ============================================================================++@ws_handler()+async def handle_get_provider_defaults(data: Dict[str, Any], websocket: WebSocket) -> Dict[str, Any]:+ """Get default parameters for a provider."""+ database = container.database()+ provider = data.get("provider", "").lower()+ defaults = await database.get_provider_defaults(provider)++ # Return defaults or standard fallback values+ return {+ "provider": provider,+ "defaults": defaults or {+ "temperature": 0.7,+ "max_tokens": 1000,+ "thinking_enabled": False,+ "thinking_budget": 2048,+ "reasoning_effort": "medium",+ "reasoning_format": "parsed",+ }+ }+++@ws_handler()+async def handle_save_provider_defaults(data: Dict[str, Any], websocket: WebSocket) -> Dict[str, Any]:+ """Save default parameters for a provider."""+ database = container.database()+ provider = data.get("provider", "").lower()+ defaults = data.get("defaults", {})+ success = await database.save_provider_defaults(provider, defaults)+ return {"success": success, "provider": provider}++# ============================================================================# Compaction Handlers# ============================================================================@@ -2165,6 +2188,10 @@ MESSAGE_HANDLERS: Dict[str, MessageHandler] = {"get_user_settings": handle_get_user_settings,"save_user_settings": handle_save_user_settings,+ # Provider Defaults+ "get_provider_defaults": handle_get_provider_defaults,+ "save_provider_defaults": handle_save_provider_defaults,+# Compaction"get_compaction_stats": handle_get_compaction_stats,"configure_compaction": handle_configure_compaction,diff --git a/server/services/ai.py b/server/services/ai.pyindex 36ffc455..97940e7c 100644--- a/server/services/ai.py+++ b/server/services/ai.py@@ -2323,6 +2323,8 @@ class AIService:'productivity_agent': 'delegate_to_productivity_agent','payments_agent': 'delegate_to_payments_agent','consumer_agent': 'delegate_to_consumer_agent',+ 'autonomous_agent': 'delegate_to_autonomous_agent',+ 'orchestrator_agent': 'delegate_to_orchestrator_agent',# Android service nodes (direct tool usage)'batteryMonitor': 'android_battery','networkMonitor': 'android_network',@@ -2370,6 +2372,8 @@ class AIService:'productivity_agent': 'ONE-SHOT delegation to Productivity Agent. Call ONCE per task, returns task_id. Agent works in background - do NOT re-call.','payments_agent': 'ONE-SHOT delegation to Payments Agent. Call ONCE per task, returns task_id. Agent works in background - do NOT re-call.','consumer_agent': 'ONE-SHOT delegation to Consumer Agent. Call ONCE per task, returns task_id. Agent works in background - do NOT re-call.',+ 'autonomous_agent': 'ONE-SHOT delegation to Autonomous Agent. Call ONCE per task, returns task_id. Agent works in background using Code Mode patterns - do NOT re-call.',+ 'orchestrator_agent': 'ONE-SHOT delegation to Orchestrator Agent. Call ONCE per task, returns task_id. Coordinates multiple agents - do NOT re-call.',# Android service nodes (direct tool usage)'batteryMonitor': 'Monitor Android battery status, level, charging state, temperature, and health.','networkMonitor': 'Monitor Android network connectivity, type, and internet availability.',@@ -2929,7 +2933,7 @@ class AIService:return CheckDelegatedTasksSchema# AI Agent delegation schema (fire-and-forget async delegation)- if node_type in ('aiAgent', 'chatAgent', 'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent'):+ if node_type in ('aiAgent', 'chatAgent', 'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent', 'autonomous_agent', 'orchestrator_agent'):agent_label = params.get('label', node_type)class DelegateToAgentSchema(BaseModel):diff --git a/server/services/android/client.py b/server/services/android/client.pyindex 678b1950..8cae4a94 100644--- a/server/services/android/client.py+++ b/server/services/android/client.py@@ -119,7 +119,7 @@ class RelayWebSocketClient:# Broadcast QR data to frontendif self.qr_data:- await broadcast_qr_code(self.qr_data)+ await broadcast_qr_code(self.qr_data, self.session_token)# Start background tasksself._receive_task = asyncio.create_task(self._receive_loop())@@ -329,7 +329,7 @@ class RelayWebSocketClient:self.session_token = params.get("session_token")self.qr_data = params.get("qr_data")if self.qr_data:- await broadcast_qr_code(self.qr_data)+ await broadcast_qr_code(self.qr_data, self.session_token)async def _handle_pairing_connected(self, params: dict):"""Handle pairing.connected event."""diff --git a/server/services/compaction.py b/server/services/compaction.pyindex 3ec09ea1..443709f6 100644--- a/server/services/compaction.py+++ b/server/services/compaction.py@@ -18,6 +18,35 @@ if TYPE_CHECKING:logger = get_logger(__name__)+def _extract_text_from_response(content) -> str:+ """Extract text content from various LLM response formats.++ Handles:+ - String content (OpenAI, Anthropic)+ - List of content blocks (Gemini format: [{"type": "text", "text": "..."}])+ - Other complex formats+ """+ # Handle list content (Gemini format)+ if isinstance(content, list):+ text_parts = []+ for block in content:+ if isinstance(block, dict):+ if block.get('type') == 'text' and block.get('text'):+ text_parts.append(block['text'])+ elif 'text' in block:+ text_parts.append(str(block['text']))+ elif isinstance(block, str):+ text_parts.append(block)+ return '\n'.join(text_parts)++ # Handle string content+ if isinstance(content, str):+ return content++ # Fallback+ return str(content) if content else ""++class CompactionConfig(BaseModel):"""Provider-agnostic compaction configuration."""enabled: bool = True@@ -161,7 +190,8 @@ Provide a concise but complete summary."""from langchain_core.messages import HumanMessageresponse = await llm.ainvoke([HumanMessage(content=prompt)])- summary = response.content if hasattr(response, 'content') else str(response)+ raw_content = response.content if hasattr(response, 'content') else str(response)+ summary = _extract_text_from_response(raw_content)new_memory = f"# Conversation Summary (Compacted)\n*Generated: {datetime.now(timezone.utc).isoformat()}*\n\n{summary}"diff --git a/server/services/handlers/ai.py b/server/services/handlers/ai.pyindex 33d08d25..413c59b3 100644--- a/server/services/handlers/ai.py+++ b/server/services/handlers/ai.py@@ -79,7 +79,13 @@ async def _collect_agent_connections(if target_handle == 'input-memory':if source_node.get('type') == 'simpleMemory':memory_params = await database.get_node_parameters(source_node_id) or {}- memory_session_id = memory_params.get('sessionId', 'default')+ # Auto-derive session from connected AI Agent's node_id+ # Memory node's sessionId is used only if explicitly set (not empty or 'default')+ configured_session = memory_params.get('sessionId', '')+ if configured_session and configured_session != 'default':+ memory_session_id = configured_session # Explicit override+ else:+ memory_session_id = node_id # Auto: use AI Agent's node_idwindow_size = int(memory_params.get('windowSize', 10))memory_content = memory_params.get('memoryContent', '# Conversation History\n\n*No messages yet.*\n')long_term_enabled = memory_params.get('longTermEnabled', False)@@ -93,7 +99,7 @@ async def _collect_agent_connections('long_term_enabled': long_term_enabled,'retrieval_count': retrieval_count}- logger.debug(f"{log_prefix} Connected memory node: session={memory_session_id}, content_length={len(memory_content)}")+ logger.debug(f"{log_prefix} Connected memory node: session={memory_session_id} (auto={not configured_session or configured_session == 'default'}), content_length={len(memory_content)}")# Skill detection - nodes connected to input-skill handleelif target_handle == 'input-skill':diff --git a/server/services/handlers/tools.py b/server/services/handlers/tools.pyindex 47eec2a6..4fa38973 100644--- a/server/services/handlers/tools.py+++ b/server/services/handlers/tools.py@@ -133,8 +133,8 @@ async def execute_tool(tool_name: str, tool_args: Dict[str, Any],return await _execute_check_delegated_tasks(tool_args, config)# AI Agent delegation (fire-and-forget async delegation)- # Includes specialized agents: android_agent, coding_agent, web_agent, task_agent, social_agent- if node_type in ('aiAgent', 'chatAgent', 'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent'):+ # Includes specialized agents: android_agent, coding_agent, web_agent, task_agent, social_agent, autonomous_agent+ if node_type in ('aiAgent', 'chatAgent', 'android_agent', 'coding_agent', 'web_agent', 'task_agent', 'social_agent', 'travel_agent', 'tool_agent', 'productivity_agent', 'payments_agent', 'consumer_agent', 'autonomous_agent', 'orchestrator_agent'):return await _execute_delegated_agent(tool_args, config)# Generic fallback for unknown node typesdiff --git a/server/services/node_executor.py b/server/services/node_executor.pyindex b9ac1c7a..518c993f 100644--- a/server/services/node_executor.py+++ b/server/services/node_executor.py@@ -120,6 +120,8 @@ class NodeExecutor:'productivity_agent': partial(handle_chat_agent, ai_service=self.ai_service, database=self.database),'payments_agent': partial(handle_chat_agent, ai_service=self.ai_service, database=self.database),'consumer_agent': partial(handle_chat_agent, ai_service=self.ai_service, database=self.database),+ 'autonomous_agent': partial(handle_chat_agent, ai_service=self.ai_service, database=self.database),+ 'orchestrator_agent': partial(handle_chat_agent, ai_service=self.ai_service, database=self.database),'simpleMemory': handle_simple_memory,# Maps'gmaps_create': partial(handle_create_map, maps_service=self.maps_service),diff --git a/server/services/skill_loader.py b/server/services/skill_loader.pyindex bb67bba0..1b4cb8c5 100644--- a/server/services/skill_loader.py+++ b/server/services/skill_loader.py@@ -184,9 +184,9 @@ class SkillLoader:if name in self._cache:return self._cache[name]- # Check registry+ # Check registry (filesystem skills only - use load_skill_async for user skills)if name not in self._registry:- logger.warning(f"[SkillLoader] Skill not found: {name}")+ logger.debug(f"[SkillLoader] Skill not in registry: {name}")return Nonemetadata = self._registry[name]@@ -275,20 +275,24 @@ class SkillLoader:# Load from databaseif self._database:try:- user_skill = await self._database.get_user_skill_by_name(name)+ user_skill = await self._database.get_user_skill(name)if user_skill:- allowed_tools = []- if user_skill.allowed_tools:- allowed_tools = [t.strip() for t in user_skill.allowed_tools.split(',')]+ # user_skill is a dict from _skill_to_dict()+ allowed_tools = user_skill.get('allowed_tools', [])+ if isinstance(allowed_tools, str):+ allowed_tools = [t.strip() for t in allowed_tools.split(',') if t.strip()]- metadata_dict = {}- if user_skill.metadata_json:+ metadata_dict = user_skill.get('metadata', {})+ if isinstance(metadata_dict, str):import json- metadata_dict = json.loads(user_skill.metadata_json)+ try:+ metadata_dict = json.loads(metadata_dict) if metadata_dict else {}+ except json.JSONDecodeError:+ metadata_dict = {}metadata = SkillMetadata(- name=user_skill.name,- description=user_skill.description,+ name=user_skill.get('name', name),+ description=user_skill.get('description', ''),allowed_tools=allowed_tools,metadata=metadata_dict,path=None@@ -296,11 +300,12 @@ class SkillLoader:skill = Skill(metadata=metadata,- instructions=user_skill.instructions,+ instructions=user_skill.get('instructions', ''),scripts={},references={})+ logger.info(f"[SkillLoader] Loaded user skill from database: {name}")self._cache[name] = skillreturn skillexcept Exception as e:diff --git a/server/services/temporal/__init__.py b/server/services/temporal/__init__.pyindex 9cd29db2..757c23c4 100644--- a/server/services/temporal/__init__.py+++ b/server/services/temporal/__init__.py@@ -15,9 +15,19 @@ When TEMPORAL_ENABLED=true:When TEMPORAL_ENABLED=false (default):- Falls back to the existing parallel/sequential executor-"""-from .executor import TemporalExecutor-from .client import TemporalClientWrapper+Note: temporalio is an optional dependency. Import with try/except.+"""__all__ = ["TemporalExecutor", "TemporalClientWrapper"]++# Lazy imports - temporalio is optional+TemporalExecutor = None+TemporalClientWrapper = None++try:+ from .executor import TemporalExecutor+ from .client import TemporalClientWrapper+except ImportError:+ # temporalio not installed - Temporal features disabled+ passdiff --git a/server/skills/assistant/compaction-skill/SKILL.md b/server/skills/assistant/compaction-skill/SKILL.mdnew file mode 100644index 00000000..55ebe0a1--- /dev/null+++ b/server/skills/assistant/compaction-skill/SKILL.md@@ -0,0 +1,157 @@+---+name: compaction-skill+description: Manage memory compaction by summarizing conversation history when approaching token limits+allowed-tools: memory-save memory-get+metadata:+ author: machina+ version: "1.0"+ category: memory+ icon: "📦"+ color: "#8B5CF6"+---++# Memory Compaction Skill++You have the ability to compact conversation memory when it grows too large. Compaction transforms verbose conversation history into a structured summary that preserves essential context while reducing token usage.++## When to Compact++Compact memory when:+- The system indicates token threshold is approaching+- Conversation history becomes repetitive or verbose+- You need to preserve important context but reduce size+- Starting a new phase of work after completing a major task++## Compaction Summary Structure++When compacting, create a summary with these 5 sections:++### 1. Task Overview+What the user is trying to accomplish. Include:+- Primary goal or objective+- Key constraints or requirements+- Scope of the work++### 2. Current State+What's been completed and what's in progress:+- Completed tasks and their outcomes+- Work currently in flight+- Pending decisions or blockers++### 3. Important Discoveries+Key findings, decisions, or problems encountered:+- Technical discoveries or insights+- Decisions made and their rationale+- Problems encountered and solutions applied+- User preferences learned++### 4. Next Steps+What needs to happen next:+- Immediate actions required+- Planned approach for remaining work+- Dependencies or prerequisites++### 5. Context to Preserve+Critical details that must be retained:+- Specific values, IDs, or references+- User preferences or constraints+- Technical details needed for continuity+- Any warnings or caveats++## Compaction Format++Output the compacted summary in this format:++```markdown+# Conversation Summary (Compacted)+*Generated: [ISO timestamp]*++## Task Overview+[1-3 sentences describing the goal]++## Current State+- [Completed item 1]+- [Completed item 2]+- [In progress: description]++## Important Discoveries+- [Discovery 1 with context]+- [Decision made: rationale]+- [Problem solved: approach]++## Next Steps+1. [Next action]+2. [Following action]++## Context to Preserve+- [Critical detail 1]+- [Critical detail 2]+```++## Best Practices++1. **Be Concise**: Each section should be brief but complete+2. **Preserve Specifics**: Keep exact values, names, and references+3. **Capture Decisions**: Record WHY decisions were made, not just WHAT+4. **Include Failures**: Document what didn't work to avoid repetition+5. **Maintain Continuity**: Summary should allow seamless continuation++## What NOT to Include++- Verbose back-and-forth dialogue+- Redundant information+- Superseded decisions (only keep final decisions)+- Exploratory tangents that didn't lead anywhere+- Standard pleasantries or acknowledgments++## Example Compaction++**Before** (verbose history):+```+Human: Can you help me debug this Python function?+AI: Of course! Please share the function.+Human: Here's the function: def calculate(x): return x * 2+AI: I see the function. What issue are you experiencing?+Human: It returns None sometimes+AI: That's interesting. Can you show me an example input?+Human: calculate("5") returns None+AI: Ah, I see the issue! When you pass a string...+[continues for 50+ messages]+```++**After** (compacted):+```markdown+# Conversation Summary (Compacted)+*Generated: 2025-02-13T10:30:00Z*++## Task Overview+Debug Python function `calculate(x)` that returns None for some inputs.++## Current State+- Identified root cause: string inputs cause implicit None return+- Implemented fix with type checking and conversion+- Tests passing for int, float, and string inputs++## Important Discoveries+- Original function had no type validation+- String multiplication in Python doesn't raise error but behaves unexpectedly+- User prefers explicit error messages over silent failures++## Next Steps+1. Add input validation for edge cases (None, empty string)+2. Write unit tests for the fixed function++## Context to Preserve+- Function location: `utils/math_helpers.py:45`+- User wants to maintain backward compatibility+- Prefer raising ValueError over returning None+```++## Integration with Memory System++When compaction is triggered:+1. The compacted summary replaces the current memory content+2. New conversation messages are appended after the summary+3. The summary header indicates when compaction occurred++This allows conversation to continue naturally while maintaining reduced token usage.diff --git a/server/skills/assistant/humanify-skill/SKILL.md b/server/skills/assistant/humanify-skill/SKILL.mdnew file mode 100644index 00000000..47bd446c--- /dev/null+++ b/server/skills/assistant/humanify-skill/SKILL.md@@ -0,0 +1,210 @@+---+name: humanify-skill+description: Output human-readable text in pretty formats instead of raw markdown or code blocks+metadata:+ author: machina+ version: "1.0"+ category: formatting+ icon: "✨"+ color: "#F59E0B"+---++# Humanify Output Skill++Format your responses for human readability. Output clean, pretty text that's easy to scan and understand - not raw markdown syntax or code-heavy formatting.++## Core Principles++1. **Plain Language First**: Write naturally, as if speaking to someone+2. **Visual Clarity**: Use spacing, indentation, and structure for scannability+3. **No Raw Syntax**: Avoid showing markdown symbols, JSON, or code unless specifically requested+4. **Progressive Detail**: Lead with key information, details follow++## Formatting Guidelines++### Instead of Markdown Headers+```+Bad: ## Section Title+Good: SECTION TITLE+ -------------+```++### Instead of Bullet Lists with Symbols+```+Bad: - Item one+ - Item two++Good: Item one+ Item two++ or: * Item one+ * Item two+```++### Instead of Code Blocks for Data+```+Bad: ```json+ {"name": "John", "age": 30}+ ```++Good: Name: John+ Age: 30+```++### Instead of Tables with Pipes+```+Bad: | Name | Value |+ |------|-------|+ | Foo | 123 |++Good: Name Value+ ---- -----+ Foo 123+```++### Instead of Raw URLs+```+Bad: Check https://example.com/very/long/path/to/resource++Good: Check the resource page (link available)+ or: Resource: example.com+```++## Response Patterns++### For Status Updates+```+Status: Complete+Time: 2.3 seconds+Result: Successfully processed 15 items++Details:+ Processed: 15+ Skipped: 2+ Errors: 0+```++### For Lists of Items+```+Found 3 matching files:++ 1. config.json+ Location: /app/settings+ Modified: Today++ 2. config.backup.json+ Location: /app/backups+ Modified: Yesterday++ 3. config.old.json+ Location: /archive+ Modified: Last week+```++### For Explanations+```+The function calculates the total by:++ First, it gathers all input values+ Then, it filters out any invalid entries+ Finally, it sums the remaining numbers++Note: Empty inputs are treated as zero.+```++### For Errors or Warnings+```+Could not complete the request.++Reason: The file was not found at the specified location.++Suggestions:+ Check if the file path is correct+ Verify the file exists+ Ensure you have read permissions+```++### For Confirmations+```+Done! Created new user account.++ Username: jsmith+ Email: [email protected]+ Role: Editor++Next: The user will receive a welcome email shortly.+```++## When to Use Each Style++### Use Pretty Formatting When:+- Showing results or status+- Explaining concepts+- Listing options or items+- Providing instructions+- Displaying data summaries++### Keep Technical Formatting When:+- User explicitly requests code/markdown+- Showing code that needs to be copied+- Displaying configuration files+- Technical documentation+- API responses that need exact format++## Typography Tips++### Visual Separators+```+Section breaks: ---------------+Light dividers: . . . . . . . .+Box drawing: +-------------++ | Content |+ +-------------++```++### Emphasis Without Markdown+```+Instead of **bold**: Use CAPS or spacing+Instead of *italic*: Use quotes or context+Instead of `code`: Use clear naming+```++### Spacing for Clarity+```+Group related items with spacing:++ Primary Actions+ Save changes+ Submit form++ Secondary Actions+ Cancel+ Reset+```++## Examples++### Bad (Raw Markdown)+```+### Results+- **Found**: 5 files+- **Size**: 2.3MB+- **Status**: `complete`++See [documentation](https://docs.example.com) for more info.+```++### Good (Humanified)+```+RESULTS+-------+Found: 5 files+Size: 2.3 MB+Status: Complete++See the documentation for more details.+```++## Remember++The goal is communication, not formatting. Choose the clearest way to convey information to a human reader. When in doubt, write it as you would explain it to a colleague standing next to you.