{"repo": "agentdock/agentdock", "files": {".env.example": {"4d9eb73747deda96eb23aea4755e160d8d482eba": {"t": 1785747422.768406, "n": 15312, "h": "7b8f412c7ab71eb6", "c": "# AgentDock Comprehensive Environment Configuration\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'memory': Default in-memory storage (no persistence).\n# - 'redis': Redis-compatible storage via Upstash client.\n# - 'vercel-kv': Vercel KV storage (Redis under the hood).\n# - 'sqlite': SQLite database (auto-enabled in development).\n# - 'postgresql': PostgreSQL database (auto-enabled when DATABASE_URL is set).\nKV_STORE_PROVIDER=memory\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# Additional Redis settings for Upstash or custom Redis instances\nUPSTASH_REDIS_REST_URL= # Upstash Redis REST endpoint\nUPSTASH_REDIS_REST_TOKEN= # Upstash Redis authentication token\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# AGENTDOCK CORE - DATABASE & MEMORY CONFIGURATION\n# ==============================================================================\n# AgentDock Core database, storage, and memory system configuration.\n\n# Database Configuration\nDATABASE_URL=postgresql://postgres:password@localhost:5432/agentdock\nPOSTGRES_HOST=localhost # PostgreSQL host\nPOSTGRES_PORT=5432 # PostgreSQL port\nPOSTGRES_DB=agentdock # Database name\nPOSTGRES_USER=postgres # Database user\nPOSTGRES_PASSWORD= # Database password\nPOSTGRES_SCHEMA=public # Schema name\n\n# Core Storage Adapters\nENABLE_SQLITE=true # Set to 'true' to force enable SQLite\nENABLE_SQLITE_VEC=false # Set to 'true' to enable vector search capabilities\nENABLE_PGVECTOR=true # Set to 'true' for vector search support\nSQLITE_PATH=./agentdock.db # Custom database path (default: ./agentdock.db)\n\n# Memory System Configuration\nAGENTDOCK_STORAGE_TYPE=postgresql # Options: postgresql, sqlite, sqlite-vec, postgresql-vector\nAGENTDOCK_SQLITE_PATH=./agentdock-memory.db # SQLite path for memory operations\nAGENTDOCK_ENABLE_SQLITE_VEC=true # Enable vector operations for memory\nAGENTDOCK_ENABLE_PGVECTOR=true # Enable PostgreSQL vector for memory\n\n# Memory Recall Configuration\nAGENTDOCK_RECALL_PRESET=default # Options: default, precision, performance, research\nMEMORY_MAX_TOKENS=10000 # Maximum tokens for memory context\nMEMORY_TTL=86400 # Memory time-to-live in seconds\nMEMORY_ENCRYPTION_KEY= # Encryption key for sensitive memories\n\n# PRIME (Priority Rules Intelligent Memory Extraction) System - 2-TIER\n# ------------------------------------------------------------------------------\n# 🚀 SIMPLEST SETUP - Just add your API key:\n# OPENAI_API_KEY=sk-xxx # That's it! Smart 2-tier selection works automatically\n#\n# 📦 ONE MODEL EVERYWHERE:\n# PRIME_MODEL=gpt-4o-mini # Force single model for both tiers\n#\n# 🎯 ADVANCED CONTROL:\n# PRIME_STANDARD_MODEL=gpt-4o-mini # Cheap model for simple content\n# PRIME_ADVANCED_MODEL=gpt-4o # Quality model for complex content\n# PRIME_DEFAULT_TIER=advanced # Force always use advanced\n#\nPRIME_PROVIDER=openai # LLM provider (openai, anthropic, gemini)\nPRIME_API_KEY= # Dedicated API key (falls back to provider key)\nPRIME_MODEL= # Override BOTH tiers with single model\nPRIME_STANDARD_MODEL= # Standard tier model (default: gpt-4o-mini)\nPRIME_ADVANCED_MODEL= # Advanced tier model (default: gpt-4o)\nPRIME_DEFAULT_TIER=standard # Default tier when auto-selection disabled\nPRIME_MAX_TOKENS=4000 # Maximum tokens for extraction\nPRIME_ADVANCED_MIN_CHARS=500 # Use advanced tier for content >N chars\nPRIME_ADVANCED_MIN_RULES=5 # Use advanced tier for >N active rules\nPRIME_ENABLE_COST_TRACKING=true # Enable extraction cost tracking\nPRIME_COST_THRESHOLD=10.00 # Daily cost threshold in USD\n\n# Memory Connection System Configuration - 2-TIER \n# ------------------------------------------------------------------------------\n# 🚀 SIMPLEST SETUP - Inherits from PRIME:\n# # No config needed! Uses PRIME settings automatically\n#\n# 📦 ONE MODEL EVERYWHERE:\n# CONNECTION_MODEL=gpt-4o-mini # Force single model for both tiers\n#\n# 🎯 ADVANCED CONTROL:\n# CONNECTION_STANDARD_MODEL=gpt-4o-mini # For normal connections\n# CONNECTION_ENHANCED_MODEL=gpt-4o # For important connections\n# CONNECTION_ALWAYS_ADVANCED=true # Force always use advanced\n#\n# Smart triage: 65% connections FREE (auto-classified), 35% use LLM\n#\nCONNECTION_PROVIDER= # Override provider (defaults to PRIME_PROVIDER)\nCONNECTION_API_KEY= # Override API key (defaults to provider key)\nCONNECTION_MODEL= # Override BOTH tiers with single model\nCONNECTION_STANDARD_MODEL= # Standard tier (default: gpt-4o-mini)\nCONNECTION_ENHANCED_MODEL= # Advanced tier (default: gpt-4o)\nCONNECTION_ADVANCED_MODEL= # Alias for CONNECTION_ENHANCED_MODEL\nCONNECTION_ALWAYS_ADVANCED=false # Force advanced model for all connections\nCONNECTION_PREFER_QUALITY=false # Bias toward quality in production\n\n# Smart triage thresholds - Control cost optimization\nCONNECTION_AUTO_SIMILAR=0.8 # Similarity >N = auto \"similar\" (40% FREE)\nCONNECTION_AUTO_RELATED=0.6 # Similarity >N = auto \"related\" (25% FREE) \nCONNECTION_LLM_REQUIRED=0.3 # Similarity >N = needs LLM (35% PAID)\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\nCEREBRAS_API_KEY= # Cerebras API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'vercel' when deployed on Vercel, 'local' or unset otherwise.\n# This helps agentdock-core detect if Vercel-specific features like `waitUntil` might be available.\nAGENTDOCK_EXECUTION_ENV=local\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none\n\n# ==============================================================================\n# ANALYTICS CONFIGURATION (PostHog)\n# ==============================================================================\n# Optional: Configure PostHog analytics for usage tracking.\n# See docs/analytics.md for more details.\n\n# Your PostHog Project API Key (found in project settings).\n# Required to enable any analytics tracking.\nNEXT_PUBLIC_POSTHOG_API_KEY=\n\n# The API host for your PostHog instance.\n# Defaults to PostHog Cloud US (https://us.i.posthog.com) if unset.\n# Use your own instance URL if self-hosting (e.g., https://posthog.example.com).\nNEXT_PUBLIC_POSTHOG_HOST=\n\n# Explicitly enable/disable analytics.\n# Set to 'true' to enable. \n# If unset, defaults to 'true' ONLY when NODE_ENV is 'production', otherwise 'false'.\nNEXT_PUBLIC_ANALYTICS_ENABLED=\n\n# ==============================================================================\n# ADVANCED STORAGE CONFIGURATION\n# ==============================================================================\n# AgentDock supports 13 storage adapters. The abstraction layer is complete,\n# but features like server-side chat persistence and AI memory are NOT yet built.\n#\n# Core Adapters (Always Available):\n# - memory, redis, vercel-kv\n#\n# Auto-Registered by App:\n# - sqlite, sqlite-vec, postgresql, postgresql-vector\n\n# ------------------------------------------------------------------------------\n# OPTIONAL STORAGE ADAPTERS (Manual Registration Required)\n# ------------------------------------------------------------------------------\n# These adapters must be manually registered in your API routes.\n# They are NOT auto-registered to keep build size small.\n\n# MongoDB Configuration\nENABLE_MONGODB=false # Set to 'true' if using MongoDB\nMONGODB_URI= # mongodb://localhost:27017/agentdock\n\n# AWS S3 Configuration\nENABLE_S3=false # Set to 'true' if using S3\nS3_BUCKET= # Your S3 bucket name\nAWS_REGION=us-east-1 # AWS region\nAWS_ACCESS_KEY_ID= # AWS credentials (or use IAM roles)\nAWS_SECRET_ACCESS_KEY=\n# AWS_SESSION_TOKEN= # Optional: For temporary credentials\n\n# AWS DynamoDB Configuration\nENABLE_DYNAMODB=false # Set to 'true' if using DynamoDB\nDYNAMODB_TABLE_NAME= # Your DynamoDB table name\n\n# Cloudflare Storage Configuration\nENABLE_CLOUDFLARE=false # Set to 'true' if using Cloudflare\n# Note: CF_KV_NAMESPACE and CF_D1_DATABASE must be bound in wrangler.toml\n\n# Vector Database Configuration\nENABLE_VECTOR_DBS=false # Set to 'true' if using vector databases\n\n# Pinecone Configuration\nPINECONE_API_KEY= # Your Pinecone API key\nPINECONE_INDEX= # Default index name\n\n# Qdrant Configuration\nQDRANT_HOST=localhost # Qdrant host\nQDRANT_PORT=6333 # Qdrant port\nQDRANT_API_KEY= # Optional: API key for Qdrant Cloud\n\n# ChromaDB Configuration\nCHROMADB_HOST=http://localhost:8000 # ChromaDB server URL\nCHROMADB_AUTH_TOKEN= # Optional: Authentication token\n\n# ==============================================================================\n# DEVELOPMENT & DEBUGGING\n# ==============================================================================\n# Development-specific configuration and debugging options.\n\nLOG_LEVEL=info # Logging level: debug, info, warn, error\nDEBUG_MEMORY=false # Enable memory system debug logging\nDEBUG_STORAGE=false # Enable storage adapter debug logging "}, "7d03f45d19e08448ec727b4c1b40c19e217e7f5a": {"t": 1785773198.329513, "n": 15312, "h": "7b8f412c7ab71eb6", "c": "# AgentDock Comprehensive Environment Configuration\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'memory': Default in-memory storage (no persistence).\n# - 'redis': Redis-compatible storage via Upstash client.\n# - 'vercel-kv': Vercel KV storage (Redis under the hood).\n# - 'sqlite': SQLite database (auto-enabled in development).\n# - 'postgresql': PostgreSQL database (auto-enabled when DATABASE_URL is set).\nKV_STORE_PROVIDER=memory\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# Additional Redis settings for Upstash or custom Redis instances\nUPSTASH_REDIS_REST_URL= # Upstash Redis REST endpoint\nUPSTASH_REDIS_REST_TOKEN= # Upstash Redis authentication token\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# AGENTDOCK CORE - DATABASE & MEMORY CONFIGURATION\n# ==============================================================================\n# AgentDock Core database, storage, and memory system configuration.\n\n# Database Configuration\nDATABASE_URL=postgresql://postgres:password@localhost:5432/agentdock\nPOSTGRES_HOST=localhost # PostgreSQL host\nPOSTGRES_PORT=5432 # PostgreSQL port\nPOSTGRES_DB=agentdock # Database name\nPOSTGRES_USER=postgres # Database user\nPOSTGRES_PASSWORD= # Database password\nPOSTGRES_SCHEMA=public # Schema name\n\n# Core Storage Adapters\nENABLE_SQLITE=true # Set to 'true' to force enable SQLite\nENABLE_SQLITE_VEC=false # Set to 'true' to enable vector search capabilities\nENABLE_PGVECTOR=true # Set to 'true' for vector search support\nSQLITE_PATH=./agentdock.db # Custom database path (default: ./agentdock.db)\n\n# Memory System Configuration\nAGENTDOCK_STORAGE_TYPE=postgresql # Options: postgresql, sqlite, sqlite-vec, postgresql-vector\nAGENTDOCK_SQLITE_PATH=./agentdock-memory.db # SQLite path for memory operations\nAGENTDOCK_ENABLE_SQLITE_VEC=true # Enable vector operations for memory\nAGENTDOCK_ENABLE_PGVECTOR=true # Enable PostgreSQL vector for memory\n\n# Memory Recall Configuration\nAGENTDOCK_RECALL_PRESET=default # Options: default, precision, performance, research\nMEMORY_MAX_TOKENS=10000 # Maximum tokens for memory context\nMEMORY_TTL=86400 # Memory time-to-live in seconds\nMEMORY_ENCRYPTION_KEY= # Encryption key for sensitive memories\n\n# PRIME (Priority Rules Intelligent Memory Extraction) System - 2-TIER\n# ------------------------------------------------------------------------------\n# 🚀 SIMPLEST SETUP - Just add your API key:\n# OPENAI_API_KEY=sk-xxx # That's it! Smart 2-tier selection works automatically\n#\n# 📦 ONE MODEL EVERYWHERE:\n# PRIME_MODEL=gpt-4o-mini # Force single model for both tiers\n#\n# 🎯 ADVANCED CONTROL:\n# PRIME_STANDARD_MODEL=gpt-4o-mini # Cheap model for simple content\n# PRIME_ADVANCED_MODEL=gpt-4o # Quality model for complex content\n# PRIME_DEFAULT_TIER=advanced # Force always use advanced\n#\nPRIME_PROVIDER=openai # LLM provider (openai, anthropic, gemini)\nPRIME_API_KEY= # Dedicated API key (falls back to provider key)\nPRIME_MODEL= # Override BOTH tiers with single model\nPRIME_STANDARD_MODEL= # Standard tier model (default: gpt-4o-mini)\nPRIME_ADVANCED_MODEL= # Advanced tier model (default: gpt-4o)\nPRIME_DEFAULT_TIER=standard # Default tier when auto-selection disabled\nPRIME_MAX_TOKENS=4000 # Maximum tokens for extraction\nPRIME_ADVANCED_MIN_CHARS=500 # Use advanced tier for content >N chars\nPRIME_ADVANCED_MIN_RULES=5 # Use advanced tier for >N active rules\nPRIME_ENABLE_COST_TRACKING=true # Enable extraction cost tracking\nPRIME_COST_THRESHOLD=10.00 # Daily cost threshold in USD\n\n# Memory Connection System Configuration - 2-TIER \n# ------------------------------------------------------------------------------\n# 🚀 SIMPLEST SETUP - Inherits from PRIME:\n# # No config needed! Uses PRIME settings automatically\n#\n# 📦 ONE MODEL EVERYWHERE:\n# CONNECTION_MODEL=gpt-4o-mini # Force single model for both tiers\n#\n# 🎯 ADVANCED CONTROL:\n# CONNECTION_STANDARD_MODEL=gpt-4o-mini # For normal connections\n# CONNECTION_ENHANCED_MODEL=gpt-4o # For important connections\n# CONNECTION_ALWAYS_ADVANCED=true # Force always use advanced\n#\n# Smart triage: 65% connections FREE (auto-classified), 35% use LLM\n#\nCONNECTION_PROVIDER= # Override provider (defaults to PRIME_PROVIDER)\nCONNECTION_API_KEY= # Override API key (defaults to provider key)\nCONNECTION_MODEL= # Override BOTH tiers with single model\nCONNECTION_STANDARD_MODEL= # Standard tier (default: gpt-4o-mini)\nCONNECTION_ENHANCED_MODEL= # Advanced tier (default: gpt-4o)\nCONNECTION_ADVANCED_MODEL= # Alias for CONNECTION_ENHANCED_MODEL\nCONNECTION_ALWAYS_ADVANCED=false # Force advanced model for all connections\nCONNECTION_PREFER_QUALITY=false # Bias toward quality in production\n\n# Smart triage thresholds - Control cost optimization\nCONNECTION_AUTO_SIMILAR=0.8 # Similarity >N = auto \"similar\" (40% FREE)\nCONNECTION_AUTO_RELATED=0.6 # Similarity >N = auto \"related\" (25% FREE) \nCONNECTION_LLM_REQUIRED=0.3 # Similarity >N = needs LLM (35% PAID)\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\nCEREBRAS_API_KEY= # Cerebras API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'vercel' when deployed on Vercel, 'local' or unset otherwise.\n# This helps agentdock-core detect if Vercel-specific features like `waitUntil` might be available.\nAGENTDOCK_EXECUTION_ENV=local\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none\n\n# ==============================================================================\n# ANALYTICS CONFIGURATION (PostHog)\n# ==============================================================================\n# Optional: Configure PostHog analytics for usage tracking.\n# See docs/analytics.md for more details.\n\n# Your PostHog Project API Key (found in project settings).\n# Required to enable any analytics tracking.\nNEXT_PUBLIC_POSTHOG_API_KEY=\n\n# The API host for your PostHog instance.\n# Defaults to PostHog Cloud US (https://us.i.posthog.com) if unset.\n# Use your own instance URL if self-hosting (e.g., https://posthog.example.com).\nNEXT_PUBLIC_POSTHOG_HOST=\n\n# Explicitly enable/disable analytics.\n# Set to 'true' to enable. \n# If unset, defaults to 'true' ONLY when NODE_ENV is 'production', otherwise 'false'.\nNEXT_PUBLIC_ANALYTICS_ENABLED=\n\n# ==============================================================================\n# ADVANCED STORAGE CONFIGURATION\n# ==============================================================================\n# AgentDock supports 13 storage adapters. The abstraction layer is complete,\n# but features like server-side chat persistence and AI memory are NOT yet built.\n#\n# Core Adapters (Always Available):\n# - memory, redis, vercel-kv\n#\n# Auto-Registered by App:\n# - sqlite, sqlite-vec, postgresql, postgresql-vector\n\n# ------------------------------------------------------------------------------\n# OPTIONAL STORAGE ADAPTERS (Manual Registration Required)\n# ------------------------------------------------------------------------------\n# These adapters must be manually registered in your API routes.\n# They are NOT auto-registered to keep build size small.\n\n# MongoDB Configuration\nENABLE_MONGODB=false # Set to 'true' if using MongoDB\nMONGODB_URI= # mongodb://localhost:27017/agentdock\n\n# AWS S3 Configuration\nENABLE_S3=false # Set to 'true' if using S3\nS3_BUCKET= # Your S3 bucket name\nAWS_REGION=us-east-1 # AWS region\nAWS_ACCESS_KEY_ID= # AWS credentials (or use IAM roles)\nAWS_SECRET_ACCESS_KEY=\n# AWS_SESSION_TOKEN= # Optional: For temporary credentials\n\n# AWS DynamoDB Configuration\nENABLE_DYNAMODB=false # Set to 'true' if using DynamoDB\nDYNAMODB_TABLE_NAME= # Your DynamoDB table name\n\n# Cloudflare Storage Configuration\nENABLE_CLOUDFLARE=false # Set to 'true' if using Cloudflare\n# Note: CF_KV_NAMESPACE and CF_D1_DATABASE must be bound in wrangler.toml\n\n# Vector Database Configuration\nENABLE_VECTOR_DBS=false # Set to 'true' if using vector databases\n\n# Pinecone Configuration\nPINECONE_API_KEY= # Your Pinecone API key\nPINECONE_INDEX= # Default index name\n\n# Qdrant Configuration\nQDRANT_HOST=localhost # Qdrant host\nQDRANT_PORT=6333 # Qdrant port\nQDRANT_API_KEY= # Optional: API key for Qdrant Cloud\n\n# ChromaDB Configuration\nCHROMADB_HOST=http://localhost:8000 # ChromaDB server URL\nCHROMADB_AUTH_TOKEN= # Optional: Authentication token\n\n# ==============================================================================\n# DEVELOPMENT & DEBUGGING\n# ==============================================================================\n# Development-specific configuration and debugging options.\n\nLOG_LEVEL=info # Logging level: debug, info, warn, error\nDEBUG_MEMORY=false # Enable memory system debug logging\nDEBUG_STORAGE=false # Enable storage adapter debug logging "}, "49e1f559dd7eab4469cbadb176dcca1b1d3b15d4": {"t": 1785773200.0221093, "n": 7512, "h": "2ed925ca4734450e", "c": "# AgentDock Environment Configuration Example\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'redis': Use for local development with Docker Compose (Redis + Proxy).\n# - 'vercel-kv': Use for Vercel deployment (via Marketplace/Upstash) OR local testing with Vercel.\n# - 'memory': Use ephemeral in-memory storage (no persistence).\nKV_STORE_PROVIDER=redis\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\nCEREBRAS_API_KEY= # Cerebras API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'vercel' when deployed on Vercel, 'local' or unset otherwise.\n# This helps agentdock-core detect if Vercel-specific features like `waitUntil` might be available.\nAGENTDOCK_EXECUTION_ENV=local\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Database Configuration (only relevant for potential Pro deployment)\n# DATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none\n\n# ==============================================================================\n# ANALYTICS CONFIGURATION (PostHog)\n# ==============================================================================\n# Optional: Configure PostHog analytics for usage tracking.\n# See docs/analytics.md for more details.\n\n# Your PostHog Project API Key (found in project settings).\n# Required to enable any analytics tracking.\nNEXT_PUBLIC_POSTHOG_API_KEY=\n\n# The API host for your PostHog instance.\n# Defaults to PostHog Cloud US (https://us.i.posthog.com) if unset.\n# Use your own instance URL if self-hosting (e.g., https://posthog.example.com).\nNEXT_PUBLIC_POSTHOG_HOST=\n\n# Explicitly enable/disable analytics.\n# Set to 'true' to enable. \n# If unset, defaults to 'true' ONLY when NODE_ENV is 'production', otherwise 'false'.\nNEXT_PUBLIC_ANALYTICS_ENABLED="}, "f7b16d5e1a11bfcad89d7dfd56c7131f8f56038e": {"t": 1785773201.3192203, "n": 7511, "h": "4ddb410ba2ed7fa0", "c": "# AgentDock Environment Configuration Example\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'redis': Use for local development with Docker Compose (Redis + Proxy).\n# - 'vercel-kv': Use for Vercel deployment (via Marketplace/Upstash) OR local testing with Vercel.\n# - 'memory': Use ephemeral in-memory storage (no persistence).\nKV_STORE_PROVIDER=redis\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\nCEREBRAS_API_KEY= #Cerebras API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'vercel' when deployed on Vercel, 'local' or unset otherwise.\n# This helps agentdock-core detect if Vercel-specific features like `waitUntil` might be available.\nAGENTDOCK_EXECUTION_ENV=local\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Database Configuration (only relevant for potential Pro deployment)\n# DATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none\n\n# ==============================================================================\n# ANALYTICS CONFIGURATION (PostHog)\n# ==============================================================================\n# Optional: Configure PostHog analytics for usage tracking.\n# See docs/analytics.md for more details.\n\n# Your PostHog Project API Key (found in project settings).\n# Required to enable any analytics tracking.\nNEXT_PUBLIC_POSTHOG_API_KEY=\n\n# The API host for your PostHog instance.\n# Defaults to PostHog Cloud US (https://us.i.posthog.com) if unset.\n# Use your own instance URL if self-hosting (e.g., https://posthog.example.com).\nNEXT_PUBLIC_POSTHOG_HOST=\n\n# Explicitly enable/disable analytics.\n# Set to 'true' to enable. \n# If unset, defaults to 'true' ONLY when NODE_ENV is 'production', otherwise 'false'.\nNEXT_PUBLIC_ANALYTICS_ENABLED="}, "5d48bbc2fc2b69dc39c768c5cd6e8a7918c5c5d3": {"t": 1785773202.5534241, "n": 7461, "h": "3a8493aae262de95", "c": "# AgentDock Environment Configuration Example\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'redis': Use for local development with Docker Compose (Redis + Proxy).\n# - 'vercel-kv': Use for Vercel deployment (via Marketplace/Upstash) OR local testing with Vercel.\n# - 'memory': Use ephemeral in-memory storage (no persistence).\nKV_STORE_PROVIDER=redis\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'vercel' when deployed on Vercel, 'local' or unset otherwise.\n# This helps agentdock-core detect if Vercel-specific features like `waitUntil` might be available.\nAGENTDOCK_EXECUTION_ENV=local\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Database Configuration (only relevant for potential Pro deployment)\n# DATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none\n\n# ==============================================================================\n# ANALYTICS CONFIGURATION (PostHog)\n# ==============================================================================\n# Optional: Configure PostHog analytics for usage tracking.\n# See docs/analytics.md for more details.\n\n# Your PostHog Project API Key (found in project settings).\n# Required to enable any analytics tracking.\nNEXT_PUBLIC_POSTHOG_API_KEY=\n\n# The API host for your PostHog instance.\n# Defaults to PostHog Cloud US (https://us.i.posthog.com) if unset.\n# Use your own instance URL if self-hosting (e.g., https://posthog.example.com).\nNEXT_PUBLIC_POSTHOG_HOST=\n\n# Explicitly enable/disable analytics.\n# Set to 'true' to enable. \n# If unset, defaults to 'true' ONLY when NODE_ENV is 'production', otherwise 'false'.\nNEXT_PUBLIC_ANALYTICS_ENABLED="}, "792bcc29a6e384d7fb6d4ebe2f9b34ea2d748b49": {"t": 1785773205.1915774, "n": 6622, "h": "21f0f0e790ec3277", "c": "# AgentDock Environment Configuration Example\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'redis': Use for local development with Docker Compose (Redis + Proxy).\n# - 'vercel-kv': Use for Vercel deployment (via Marketplace/Upstash) OR local testing with Vercel.\n# - 'memory': Use ephemeral in-memory storage (no persistence).\nKV_STORE_PROVIDER=redis\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'vercel' when deployed on Vercel, 'local' or unset otherwise.\n# This helps agentdock-core detect if Vercel-specific features like `waitUntil` might be available.\nAGENTDOCK_EXECUTION_ENV=local\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Database Configuration (only relevant for potential Pro deployment)\n# DATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none"}, "115f6819c06942d880d4d7ca8ffabae47d71db07": {"t": 1785773206.7627468, "n": 6403, "h": "849b21097e7c9aad", "c": "# AgentDock Environment Configuration Example\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# Storage Provider Selection\n# ==============================================================================\n# Select the Key-Value storage provider.\n# - 'redis': Use for local development with Docker Compose (Redis + Proxy).\n# - 'vercel-kv': Use for Vercel deployment (via Marketplace/Upstash) OR local testing with Vercel.\n# - 'memory': Use ephemeral in-memory storage (no persistence).\nKV_STORE_PROVIDER=redis\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=redis (Local Docker Compose)\n# ==============================================================================\n# These are used ONLY when KV_STORE_PROVIDER is set to 'redis'.\n# They point to the local redis-http-proxy container defined in docker-compose.yaml.\nREDIS_URL=http://localhost:8079\n# REDIS_TOKEN= # Only needed if your local Redis requires auth (default Docker setup doesn't)\nSRH_TOKEN=test_token # Required authentication token for the local redis-http-proxy.\n\n# ==============================================================================\n# Configuration for KV_STORE_PROVIDER=vercel-kv (Vercel Deployment / Local Test)\n# ==============================================================================\n# These variables are used ONLY when KV_STORE_PROVIDER is set to 'vercel-kv'.\n# When deploying to Vercel and using a Marketplace integration (like Upstash),\n# Vercel automatically injects these variables into the environment. \n# You typically DO NOT need to set them manually in Vercel project settings.\n# For local testing with `vercel dev`, these are pulled automatically after `vercel link`.\n# For local testing WITHOUT `vercel dev`, copy these values from your Vercel project's\n# Upstash integration settings into your .env.local file.\n#\n# KV_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_URL= # Automatically injected by Vercel / Set manually for local test\n# KV_REST_API_TOKEN= # Automatically injected by Vercel / Set manually for local test\n#\n# Note: Vercel also provides KV_REST_API_READ_ONLY_TOKEN and a direct REDIS_URL,\n# but these are not typically used by the default @vercel/kv setup.\n\n# ==============================================================================\n# SESSION TTL (Optional - Vercel KV / Redis)\n# ==============================================================================\n# Define how long inactive sessions should persist in seconds (e.g., 86400 for 24 hours).\nSESSION_TTL_SECONDS=86400\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n# These will be used as fallbacks if no custom key is set in settings\n# Used in priority order: 1) Agent-specific custom key, 2) Global settings key, 3) Environment variable\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\n\n# ==============================================================================\n# MESSAGE HISTORY SETTINGS\n# ==============================================================================\n# Controls conversation memory retention\n# These settings have HIGHEST PRECEDENCE and will override URL parameters\n# and agent template settings for security\n\nNEXT_PUBLIC_DEFAULT_HISTORY_POLICY=lastN # Options: none, lastN, all\nNEXT_PUBLIC_DEFAULT_HISTORY_LENGTH=20 # Number of user messages to retain\n\n# ==============================================================================\n# SCIENTIFIC API KEYS & CONFIGURATION\n# ==============================================================================\n# Required only for higher rate limits with scientific APIs\n\nPUBMED_API_KEY= # Optional: Higher rate limits for PubMed E-utilities\nOPENALEX_EMAIL= # Optional: Email for higher OpenAlex API rate limits\nSEMANTIC_SCHOLAR_API_KEY= # Optional: API key for higher Semantic Scholar rate limits\n\n# ==============================================================================\n# OTHER TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\n\n# Environment identifier: 'hub' for production/hosted, 'oss' for local/self-hosted default\nDEPLOYMENT_ENV=oss\nNEXT_PUBLIC_DEPLOYMENT_ENV=oss # Should match DEPLOYMENT_ENV\n\n# Client-side environment indicator (matches NODE_ENV by default)\nNEXT_PUBLIC_ENV=development\n\n# Database Configuration (only relevant for potential Pro deployment)\n# DATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n\n# Rate Limiting (Control via Production Env Vars on Vercel)\nENABLE_RATE_LIMIT=no # Set to 'yes' in Vercel Production Env Vars to enable\n# RATE_LIMIT_REQUESTS=20 # Set in Vercel Production Env Vars\n# RATE_LIMIT_WINDOW_SECONDS=600 # Set in Vercel Production Env Vars\n\n################################################################################\n# Services & External APIs\n################################################################################\n\n# Example Custom API\nCUSTOM_API_KEY=...\n\n################################################################################\n# App Configuration\n################################################################################\n\n# Server authentication mode (oauth, none)\nNEXTAUTH_SECRET=...\nNEXTAUTH_URL=...\nNEXTAUTH_AUTH_MODE=none"}, "1f3b53180916ac9a903008006c1c6f88fc96200f": {"t": 1785773208.1089573, "n": 1626, "h": "7b168818a3735769", "c": "# AgentDock Environment Configuration Example\n# Copy this file to .env.local for local development\n\n# ==============================================================================\n# LLM PROVIDER API KEYS\n# ==============================================================================\n# At least one of these API keys is required\n\nANTHROPIC_API_KEY= # Anthropic API key (starts with sk-ant-)\nOPENAI_API_KEY= # OpenAI API key (starts with sk-)\nGEMINI_API_KEY= # Google Gemini API key\nDEEPSEEK_API_KEY= # DeepSeek API key\nGROQ_API_KEY= # Groq API key\n\n# ==============================================================================\n# TOOL-SPECIFIC API KEYS\n# ==============================================================================\n# Required only for specific tools\n\nSERPER_API_KEY= # Required for search functionality\nFIRECRAWL_API_KEY= # Required for deeper web search\nALPHAVANTAGE_API_KEY= # Used for financial data tools\nCOINGECKO_API_KEY= # Used for cryptocurrency tools\n\n# ==============================================================================\n# DEPLOYMENT CONFIGURATION\n# ==============================================================================\n\nNODE_ENV=development # Set to 'production' for deployment\nDEPLOYMENT_TYPE=oss # Options: 'oss', 'pro'\n\n# Database Configuration (only relevant for Pro deployment)\nDATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n\n# Rate Limiting\nENABLE_RATE_LIMIT=no # Options: 'yes', 'no' "}, "c1182f390b1ebbd4db50575b8167aa8c21934261": {"t": 1785773209.3392384, "n": 1187, "h": "f3c9dae005d51f85", "c": "# Database Configuration\nDATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n# Set to 'production' for Vercel/Fly.io deployments\nNODE_ENV=development\n\n# Deployment Configuration\nDEPLOYMENT_TYPE=oss\n\n# Service Provider Configuration\n# Options: 'openrouter', 'google', 'none'\nSERVICE_PROVIDER=none\n\n# Service Provider API Keys\n# Only required if SERVICE_PROVIDER is set to the corresponding provider\nOPENROUTER_API_KEY=\nGOOGLE_API_KEY=\n\n# Other API Keys\nSERPER_API_KEY=\nFIRECRAWL_API_KEY=\nALPHAVANTAGE_API_KEY=\nCOINGECKO_API_KEY=\n\n# Rate Limiting\n# Only applies to service provider keys, not BYOK\nENABLE_RATE_LIMIT=no\n\n# Edge Function Configuration\n# MAX_DURATION: Maximum duration in seconds for Edge API functions (default: 120)\n# Vercel limits: 60s (Hobby), 300s (Pro), 900s (Enterprise)\n# MAX_DURATION=300\n\n# Note: For production deployments:\n# - Set NODE_ENV=production\n# - Configure appropriate SERVICE_PROVIDER\n# - Add corresponding API key\n# - Consider enabling rate limiting with ENABLE_RATE_LIMIT=yes\n# - Set MAX_DURATION based on your Vercel plan limits\n\n# - Vercel: Configure DATABASE_URL in the project settings\n# - Fly.io: Use `fly secrets set DATABASE_URL=...` "}, "38e13a24791a14567ab41a556f722231a4e168f3": {"t": 1785773210.686399, "n": 1146, "h": "269cc2ff5c21e0d6", "c": "# Database Configuration\nDATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n# Set to 'production' for Vercel/Fly.io deployments\nNODE_ENV=development\n\n# Deployment Configuration\nDEPLOYMENT_TYPE=oss\n\n# Service Provider Configuration\n# Options: 'openrouter', 'google', 'none'\nSERVICE_PROVIDER=none\n\n# Service Provider API Keys\n# Only required if SERVICE_PROVIDER is set to the corresponding provider\nOPENROUTER_API_KEY=\nGOOGLE_API_KEY=\n\n# Other API Keys\nSERPER_API_KEY=\nFIRECRAWL_API_KEY=\n\n# Rate Limiting\n# Only applies to service provider keys, not BYOK\nENABLE_RATE_LIMIT=no\n\n# Edge Function Configuration\n# MAX_DURATION: Maximum duration in seconds for Edge API functions (default: 120)\n# Vercel limits: 60s (Hobby), 300s (Pro), 900s (Enterprise)\n# MAX_DURATION=300\n\n# Note: For production deployments:\n# - Set NODE_ENV=production\n# - Configure appropriate SERVICE_PROVIDER\n# - Add corresponding API key\n# - Consider enabling rate limiting with ENABLE_RATE_LIMIT=yes\n# - Set MAX_DURATION based on your Vercel plan limits\n\n# - Vercel: Configure DATABASE_URL in the project settings\n# - Fly.io: Use `fly secrets set DATABASE_URL=...` "}, "48e386265a32036313b0695dfc33ab2a1d63e640": {"t": 1785773212.701301, "n": 847, "h": "07cf7e211502e1e3", "c": "# Database Configuration\nDATABASE_URL=postgresql://user:password@localhost:5432/agentdock\n# Set to 'production' for Vercel/Fly.io deployments\nNODE_ENV=development\n\n# Deployment Configuration\nDEPLOYMENT_TYPE=oss\n\n# Service Provider Configuration\n# Options: 'openrouter', 'google', 'none'\nSERVICE_PROVIDER=none\n\n# Service Provider API Keys\n# Only required if SERVICE_PROVIDER is set to the corresponding provider\nOPENROUTER_API_KEY=\nGOOGLE_API_KEY=\n\n# Rate Limiting\n# Only applies to service provider keys, not BYOK\nENABLE_RATE_LIMIT=no\n\n# Note: For production deployments:\n# - Set NODE_ENV=production\n# - Configure appropriate SERVICE_PROVIDER\n# - Add corresponding API key\n# - Consider enabling rate limiting with ENABLE_RATE_LIMIT=yes\n\n# - Vercel: Configure DATABASE_URL in the project settings\n# - Fly.io: Use `fly secrets set DATABASE_URL=...` "}}, "docs/i18n/italian/README.md": {"4d9eb73747deda96eb23aea4755e160d8d482eba": {"t": 1786068580.893615, "n": 23057, "h": "f5c72acba1e44a15", "c": "

\n \"AgentDock\n

\n\n## 🌐 Traduzioni README\n\n[Français](/docs/i18n/french/README.md) • [日本語](/docs/i18n/japanese/README.md) • [한국어](/docs/i18n/korean/README.md) • [中文](/docs/i18n/chinese/README.md) • [Español](/docs/i18n/spanish/README.md) • [Italiano](/docs/i18n/italian/README.md) • [Nederlands](/docs/i18n/dutch/README.md) • [Deutsch](/docs/i18n/deutsch/README.md) • [Polski](/docs/i18n/polish/README.md) • [Türkçe](/docs/i18n/turkish/README.md) • [Українська](/docs/i18n/ukrainian/README.md) • [Ελληνικά](/docs/i18n/greek/README.md) • [Русский](/docs/i18n/russian/README.md) • [العربية](/docs/i18n/arabic/README.md)\n\n# AgentDock: Crea Possibilità Infinite con Agenti AI\n\nAgentDock è un framework per costruire sofisticati agenti AI che svolgono compiti complessi con **determinismo configurabile**. È composto da due componenti principali:\n\n1. **AgentDock Core**: Un framework open-source, backend-first per costruire e distribuire agenti AI. È progettato per essere *agnostico rispetto al framework* e *agnostico rispetto al provider*, dandoti il controllo completo sull'implementazione del tuo agente.\n\n2. **Client Open Source**: Un'applicazione Next.js completa che serve come implementazione di riferimento e consumatore del framework AgentDock Core. Puoi vederla in azione su [https://hub.agentdock.ai](https://hub.agentdock.ai)\n\nCostruito con TypeScript, AgentDock enfatizza la *semplicità*, l'*estensibilità* e il ***determinismo configurabile***, rendendolo ideale per costruire sistemi AI affidabili e prevedibili che possono operare con una supervisione minima.\n\n## 🧠 Principi di Progettazione\n\nAgentDock è costruito su questi principi fondamentali:\n\n- **Semplicità Prima di Tutto**: Codice minimo richiesto per creare agenti funzionali\n- **Architettura Basata su Nodi**: Tutte le capacità sono implementate come nodi\n- **Strumenti come Nodi Specializzati**: Gli strumenti estendono il sistema di nodi per le capacità dell'agente\n- **Determinismo Configurabile**: Controlla la prevedibilità del comportamento dell'agente\n- **Sicurezza dei Tipi**: Tipi TypeScript completi ovunque\n\n### Determinismo Configurabile\n\nIl ***determinismo configurabile*** è una pietra miliare della filosofia di progettazione di AgentDock, permettendoti di bilanciare le capacità creative dell'AI con un comportamento di sistema prevedibile:\n\n- Gli AgentNode sono intrinsecamente non deterministici poiché gli LLM possono generare risposte diverse ogni volta\n- I workflow possono essere resi più deterministici attraverso *percorsi di esecuzione degli strumenti definiti*\n- Gli sviluppatori possono **controllare il livello di determinismo** configurando quali parti del sistema utilizzano l'inferenza LLM\n- Anche con componenti LLM, il comportamento complessivo del sistema rimane **prevedibile** attraverso interazioni strutturate degli strumenti\n- Questo approccio equilibrato consente sia la *creatività* che l'**affidabilità** nelle tue applicazioni AI\n\n#### Workflow Deterministici\n\nAgentDock supporta pienamente i workflow deterministici con cui hai familiarità dai tipici costruttori di workflow. Tutti i percorsi di esecuzione prevedibili e i risultati affidabili che ti aspetti sono disponibili, con o senza inferenza LLM:\n\n```mermaid\nflowchart LR\n Input[Input] --> Process[Processo]\n Process --> Database[(Database)]\n Process --> Output[Output]\n\n style Input fill:#f9f9f9,stroke:#333,stroke-width:1px\n style Output fill:#f9f9f9,stroke:#333,stroke-width:1px\n style Process fill:#d4f1f9,stroke:#333,stroke-width:1px\n style Database fill:#e8e8e8,stroke:#333,stroke-width:1px\n```\n\n#### Comportamento Agente Non Deterministico\n\nCon AgentDock, puoi anche sfruttare gli AgentNode con LLM quando hai bisogno di maggiore adattabilità. Gli output creativi possono variare in base alle tue esigenze, mantenendo comunque pattern di interazione strutturati:\n\n```mermaid\nflowchart TD\n Input[Query Utente] --> Agent[AgentNode]\n Agent -->|\"Ragionamento LLM (Non Deterministico)\"| ToolChoice{Selezione Strumento}\n ToolChoice -->|\"Opzione A\"| ToolA[Strumento Ricerca Approfondita]\n ToolChoice -->|\"Opzione B\"| ToolB[Strumento Analisi Dati]\n ToolChoice -->|\"Opzione C\"| ToolC[Risposta Diretta]\n ToolA --> Response[Risposta Finale]\n ToolB --> Response\n ToolC --> Response\n\n style Input fill:#f9f9f9,stroke:#333,stroke-width:1px\n style Agent fill:#ffdfba,stroke:#333,stroke-width:1px\n style ToolChoice fill:#ffdfba,stroke:#333,stroke-width:1px\n style ToolA fill:#d4f1f9,stroke:#333,stroke-width:1px\n style ToolB fill:#d4f1f9,stroke:#333,stroke-width:1px\n style ToolC fill:#d4f1f9,stroke:#333,stroke-width:1px\n style Response fill:#f9f9f9,stroke:#333,stroke-width:1px\n```\n\n#### Agenti Non Deterministici con Sotto-Workflow Deterministici\n\nAgentDock ti offre il ***meglio di entrambi i mondi*** combinando l'intelligenza dell'agente non deterministico con l'esecuzione deterministica del workflow:\n\n```mermaid\nflowchart TD\n Input[Query Utente] --> Agent[AgentNode]\n Agent -->|\"Ragionamento LLM (Non Deterministico)\"| FlowChoice{Selezione Sotto-Workflow}\n FlowChoice -->|\"Decisione A\"| Flow1[Workflow Deterministico 1]\n FlowChoice -->|\"Decisione B\"| Flow2[Workflow Deterministico 2]\n FlowChoice -->|\"Decisione C\"| DirectResponse[Genera Risposta]\n\n Flow1 --> |\"Passo 1 → 2 → 3 → ... → 200\"| Flow1Result[Risultato Workflow 1]\n Flow2 --> |\"Passo 1 → 2 → 3 → ... → 100\"| Flow2Result[Risultato Workflow 2]\n\n Flow1Result --> Response[Risposta Finale]\n Flow2Result --> Response\n DirectResponse --> Response\n\n style Input fill:#f9f9f9,stroke:#333,stroke-width:1px\n style Agent fill:#ffdfba,stroke:#333,stroke-width:1px\n style FlowChoice fill:#ffdfba,stroke:#333,stroke-width:1px\n style Flow1 fill:#c9e4ca,stroke:#333,stroke-width:1px\n style Flow2 fill:#c9e4ca,stroke:#333,stroke-width:1px\n style Flow1Result fill:#c9e4ca,stroke:#333,stroke-width:1px\n style Flow2Result fill:#c9e4ca,stroke:#333,stroke-width:1px\n style DirectResponse fill:#ffdfba,stroke:#333,stroke-width:1px\n style Response fill:#f9f9f9,stroke:#333,stroke-width:1px\n```\n\nQuesto approccio consente a workflow complessi multi-step (potenzialmente coinvolgendo centinaia di passaggi deterministici implementati all'interno di strumenti o come sequenze di nodi collegati) di essere invocati da decisioni di agenti intelligenti. Ogni workflow si esegue in modo prevedibile nonostante sia attivato da un ragionamento di agente non deterministico.\n\nPer workflow di agenti AI più avanzati e pipeline di elaborazione multi-stage, stiamo costruendo [AgentDock Pro](../../docs/agentdock-pro.md) - una potente piattaforma per creare, visualizzare ed eseguire sistemi di agenti complessi.\n\n#### In breve: il Determinismo Configurabile\n\nPensalo come quando guidi. A volte hai bisogno della creatività dell'AI (come navigare nelle strade cittadine - non deterministico), e a volte hai bisogno di processi affidabili, passo dopo passo (come seguire i segnali autostradali - deterministico). AgentDock ti consente di costruire sistemi che usano *entrambi*, scegliendo l'approccio giusto per ogni parte di un compito. Ottieni sia l'intelligenza dell'AI *che* risultati prevedibili quando ti servono.\n\n## 🏗️ Architettura Core\n\nIl framework è costruito attorno a un sistema potente e modulare basato su nodi, che funge da base per tutte le funzionalità dell'agente. Questa architettura utilizza tipi di nodi distinti come blocchi costitutivi:\n\n- **`BaseNode`**: La classe fondamentale che stabilisce l'interfaccia principale e le capacità per tutti i nodi.\n- **`AgentNode`**: Un nodo core specializzato che orchestra le interazioni LLM, l'uso degli strumenti e la logica dell'agente.\n- **Strumenti e Nodi Personalizzati**: Gli sviluppatori implementano le capacità dell'agente e la logica personalizzata come nodi che estendono `BaseNode`.\n\nQuesti nodi interagiscono tramite registri gestiti e possono essere collegati (sfruttando le porte dell'architettura core e un potenziale bus di messaggi) per abilitare comportamenti e workflow di agenti complessi, configurabili e potenzialmente deterministici.\n\nPer una spiegazione dettagliata dei componenti e delle capacità del sistema di nodi, consulta la [Documentazione del Sistema di Nodi](../../docs/nodes/README.md).\n\n## 🚀 Iniziare\n\nPer una guida completa, consulta la [Guida Introduttiva](../../docs/getting-started.md).\n\n### Requisiti\n\n* Node.js ≥ 20.11.0 (LTS)\n* pnpm ≥ 9.15.0 (Richiesto)\n* Chiavi API per provider LLM (Anthropic, OpenAI, etc.)\n\n### Installazione\n\n1. **Clona il Repository**:\n\n ```bash\n git clone https://github.com/AgentDock/AgentDock.git\n cd AgentDock\n ```\n\n2. **Installa pnpm**:\n\n ```bash\n corepack enable\n corepack prepare pnpm@latest --activate\n ```\n\n3. **Installa le Dipendenze**:\n\n ```bash\n pnpm install\n ```\n\n Per una reinstallazione pulita (quando devi ricostruire da zero):\n\n ```bash\n pnpm run clean-install\n ```\n\n Questo script rimuove tutti i node_modules, i file di lock e reinstalla correttamente le dipendenze.\n\n4. **Configura l'Ambiente**:\n\n Crea un file di ambiente (`.env` o `.env.local`) basato sul file `.env.example` fornito:\n\n ```bash\n # Opzione 1: Crea .env.local\n cp .env.example .env.local\n\n # Opzione 2: Crea .env\n cp .env.example .env\n ```\n\n Poi aggiungi le tue chiavi API al file di ambiente.\n\n5. **Avvia il Server di Sviluppo**:\n\n ```bash\n pnpm dev\n ```\n\n### Capacità Avanzate\n\n| Capacità | Descrizione | Documentazione |\n| :------------------------ | :---------------------------------------------------------------------------------------------- | :---------------------------------------------------------------------------------- |\n| **Gestione Sessioni** | Gestione dello stato isolata e performante per le conversazioni | [Documentazione Sessioni](../../docs/architecture/sessions/README.md) |\n| **Framework Orchestrazione** | Controllo del comportamento dell'agente e disponibilità degli strumenti basato sul contesto | [Documentazione Orchestrazione](../../docs/architecture/orchestration/README.md) |\n| **Astrazione Storage** | Sistema di storage flessibile con provider collegabili per KV, Vector e Secure Storage | [Documentazione Storage](../../docs/storage/README.md) |\n\nIl sistema di storage si sta attualmente evolvendo con storage chiave-valore (provider Memory, Redis, Vercel KV) e storage sicuro lato client, mentre lo storage vettoriale e backend aggiuntivi sono in fase di sviluppo.\n\n## 📕 Documentazione\n\nLa documentazione del framework AgentDock è disponibile su [hub.agentdock.ai/docs](https://hub.agentdock.ai/docs) e nella cartella `/docs/` di questo repository. La documentazione include:\n\n- Guide introduttive\n- Riferimenti API\n- Tutorial sulla creazione di nodi\n- Esempi di integrazione\n\n## 📂 Struttura Repository\n\nQuesto repository contiene:\n\n1. **AgentDock Core**: Il framework core situato in `agentdock-core/`\n2. **Client Open Source**: Un'implementazione di riferimento completa costruita con Next.js, che serve come consumatore del framework AgentDock Core.\n3. **Agenti Esempio**: Configurazioni di agenti pronte all'uso nella directory `agents/`\n\nPuoi usare AgentDock Core indipendentemente nelle tue applicazioni, o usare questo repository come punto di partenza per costruire le tue applicazioni basate su agenti.\n\n## 📝 Template Agenti\n\nAgentDock include diversi template di agenti preconfigurati. Esplorali nella directory `agents/` o leggi la [Documentazione Template Agenti](../../docs/agent-templates.md) per i dettagli di configurazione.\n\n## 🔧 Implementazioni Esempio\n\nLe implementazioni esempio mostrano casi d'uso specializzati e funzionalità avanzate:\n\n| Implementazione | Descrizione | Stato |\n| :--------------------------- | :--------------------------------------------------------------------------------------------- | :---------- |\n| **Agente Orchestrato** | Agente esempio che utilizza l'orchestrazione per adattare il comportamento in base al contesto | Disponibile |\n| **Ragionatore Cognitivo** | Affronta problemi complessi usando ragionamento strutturato e strumenti cognitivi | Disponibile |\n| **Pianificatore Agenti** | Agente specializzato per progettare e implementare altri agenti AI | Disponibile |\n| [**Playground di Codice (Code Playground)**](../../docs/roadmap/code-playground.md) | Generazione ed esecuzione di codice sandboxed con ricche capacità di visualizzazione | Pianificato |\n\n## 🔐 Dettagli Configurazione Ambiente\n\nIl Client Open Source di AgentDock richiede chiavi API per i provider LLM per funzionare. Queste sono configurate in un file di ambiente (`.env` o `.env.local`) che crei basandoti sul file `.env.example` fornito.\n\n### Chiavi API Provider LLM\n\nAggiungi le tue chiavi API del provider LLM (almeno una richiesta):\n\n```bash\n# Chiavi API Provider LLM - almeno una richiesta\nANTHROPIC_API_KEY=sk-ant-xxxxxxx # Chiave API Anthropic\nOPENAI_API_KEY=sk-xxxxxxx # Chiave API OpenAI\nGEMINI_API_KEY=xxxxxxx # Chiave API Google Gemini\nDEEPSEEK_API_KEY=xxxxxxx # Chiave API DeepSeek\nGROQ_API_KEY=xxxxxxx # Chiave API Groq\n```\n\n### Risoluzione Chiave API\n\nIl Client Open Source di AgentDock segue un ordine di priorità nel risolvere quale chiave API usare:\n\n1. **Chiave API personalizzata per agente** (impostata tramite le impostazioni dell'agente nell'UI)\n2. **Chiave API di impostazione globale** (impostata tramite la pagina delle impostazioni nell'UI)\n3. **Variabile d'ambiente** (da .env.local o piattaforma di deploy)\n\n### Chiavi API Specifiche per Strumento\n\nAlcuni strumenti richiedono anche le proprie chiavi API:\n\n```bash\n# Chiavi API Specifiche per Strumento\nSERPER_API_KEY= # Richiesto per funzionalità di ricerca\nFIRECRAWL_API_KEY= # Richiesto per ricerca web più approfondita\n```\n\nPer maggiori dettagli sulla configurazione dell'ambiente, consulta l'implementazione in [`src/types/env.ts`](../../src/types/env.ts).\n\n### Usa le Tue Chiavi API (BYOK)\n\nAgentDock segue un modello BYOK (Bring Your Own Key - Usa la Tua Chiave):\n\n1. Aggiungi le tue chiavi API nella pagina delle impostazioni dell'applicazione\n2. Alternativamente, fornisci le chiavi tramite header di richiesta per l'uso diretto dell'API\n3. Le chiavi sono memorizzate in modo sicuro usando il sistema di crittografia integrato\n4. Nessuna chiave API viene condivisa o memorizzata sui nostri server\n\n## 📦 Gestore dei Pacchetti\n\nQuesto progetto *richiede* l'uso di `pnpm` per una gestione coerente delle dipendenze. `npm` e `yarn` non sono supportati.\n\n## 💡 Cosa Puoi Costruire\n\n1. **Applicazioni Basate su AI**\n - Chatbot personalizzati con qualsiasi frontend\n - Assistenti AI da riga di comando\n - Pipeline di elaborazione dati automatizzate\n - Integrazioni di servizi backend\n\n2. **Capacità di Integrazione**\n - Qualsiasi provider AI (OpenAI, Anthropic, etc.)\n - Qualsiasi framework frontend\n - Qualsiasi servizio backend\n - Sorgenti dati e API personalizzate\n\n3. **Sistemi di Automazione**\n - Workflow di elaborazione dati\n - Pipeline di analisi documenti\n - Sistemi di reporting automatizzati\n - Agenti di automazione compiti\n\n## Caratteristiche Principali\n\n| Caratteristica | Descrizione |\n| :----------------------------- | :--------------------------------------------------------------------------------------------- |\n| 🔌 **Agnostico al Framework (Backend Node.js)** | La libreria core si integra con stack backend Node.js. |\n| 🧩 **Design Modulare** | Costruisci sistemi complessi da nodi semplici |\n| 🛠️ **Estensibile** | Crea nodi personalizzati per qualsiasi funzionalità |\n| 🔒 **Sicuro** | Funzionalità di sicurezza integrate per chiavi API e dati |\n| 🔑 **BYOK** | Usa le *tue chiavi API* per i provider LLM |\n| 📦 **Auto-sufficiente** | Il framework core ha dipendenze minime |\n| ⚙️ **Chiamate Strumento Multi-Fase (Multi-Step Tool Calls)** | Supporto per *catene di ragionamento complesse* |\n| 📊 **Logging Strutturato** | Insight dettagliati sull'esecuzione dell'agente |\n| 🛡️ **Gestione Errori Robusta** | Comportamento prevedibile e debugging semplificato |\n| 📝 **TypeScript First** | Sicurezza dei tipi ed esperienza sviluppatore migliorata |\n| 🌐 **Client Open Source** | Include implementazione di riferimento completa Next.js |\n| 🔄 **Orchestrazione** | *Controllo dinamico* del comportamento dell'agente basato sul contesto |\n| 💾 **Gestione Sessioni** | Stato isolato per conversazioni concorrenti |\n| 🎮 **Determinismo Configurabile** | Bilancia creatività AI e prevedibilità tramite logica nodi/workflow. |\n\n## 🧰 Componenti\n\nL'architettura modulare di AgentDock è costruita su questi componenti chiave:\n\n* **BaseNode**: La base per tutti i nodi nel sistema\n* **AgentNode**: L'astrazione principale per la funzionalità dell'agente\n* **Strumenti e Nodi Personalizzati**: Capacità invocabili e logica personalizzata implementate come nodi.\n* **Registro Nodi**: Gestisce la registrazione e il recupero di tutti i tipi di nodo\n* **Registro Strumenti**: Gestisce la disponibilità degli strumenti per gli agenti\n* **CoreLLM**: Interfaccia unificata per interagire con i provider LLM\n* **Registro Provider**: Gestisce le configurazioni dei provider LLM\n* **Gestione Errori**: Sistema per gestire gli errori e garantire un comportamento prevedibile\n* **Logging**: Sistema di logging strutturato per monitoraggio e debugging\n* **Orchestrazione**: Controlla la disponibilità degli strumenti e il comportamento in base al contesto della conversazione\n* **Sessioni**: Gestisce l'isolamento dello stato tra conversazioni concorrenti\n\nPer documentazione tecnica dettagliata su questi componenti, consulta la [Panoramica Architettura](../../docs/architecture/README.md).\n\n## 🗺️ Roadmap\n\nDi seguito la nostra roadmap di sviluppo per AgentDock. La maggior parte dei miglioramenti elencati qui riguarda il framework core di AgentDock (`agentdock-core`), attualmente sviluppato localmente e che sarà pubblicato come pacchetto NPM versionato una volta raggiunta una release stabile. Alcuni elementi della roadmap potrebbero anche comportare miglioramenti all'implementazione del client open-source.\n\n| Caratteristica | Descrizione | Categoria |\n| :---------------------------------------------------------------------------- | :---------------------------------------------------------------------------------------------- | :------------- |\n| [**Livello Astrazione Storage**](../../docs/roadmap/storage-abstraction.md) | Sistema di storage flessibile con provider collegabili | **In Corso** |\n| [**Sistemi Memoria Avanzati**](../../docs/roadmap/advanced-memory.md) | Gestione del contesto a lungo termine | **In Corso** |\n| [**Integrazione Storage Vettoriale**](../../docs/roadmap/vector-storage.md) | Recupero basato su embedding per documenti e memoria | **In Corso** |\n| [**Valutazione Agenti AI**](../../docs/roadmap/evaluation-framework.md) | Framework completo di testing e valutazione | **In Corso** |\n| [**Integrazione Piattaforme**](../../docs/roadmap/platform-integration.md) | Supporto per Telegram, WhatsApp e altre piattaforme di messaggistica | **Pianificato**|\n| [**Collaborazione Multi-Agente**](../../docs/roadmap/multi-agent-collaboration.md)| Consentire agli agenti di lavorare insieme | **Pianificato**|\n| [**Integrazione Protocollo Contesto Modello (MCP)**](../../docs/roadmap/mcp-integration.md) | Supporto per scoprire e usare strumenti esterni tramite MCP | **Pianificato**|\n| [**Agenti AI Vocali**](../../docs/roadmap/voice-agents.md) | Agenti AI che usano interfacce vocali e numeri di telefono tramite AgentNode | **Pianificato**|\n| [**Telemetria e Tracciabilità**](../../docs/roadmap/telemetry.md) | Logging avanzato e tracciamento delle prestazioni | **Pianificato**|\n| [**Workflow Runtime & Node Tipi**](../../docs/roadmap/workflow-nodes.md) | Runtime core, tipi di nodi e logica di orchestrazione per automazioni complesse | **Pianificato**|\n| [**AgentDock Pro**](../../docs/agentdock-pro.md) | Piattaforma cloud enterprise completa per scalare agenti AI e workflow AI | **Cloud** |\n| [**Costruttore Agenti AI in Linguaggio Naturale**](../../docs/roadmap/nl-agent-builder.md)| Costruttore visuale + costruzione di agenti e workflow in linguaggio naturale | **Cloud** |\n| [**Marketplace Agenti**](../../docs/roadmap/agent-marketplace.md) | Template agenti monetizzabili | **Cloud** |\n\n## 👥 Contribuire\n\nAccogliamo contributi ad AgentDock! Consulta [CONTRIBUTING.md](../../CONTRIBUTING.md) per linee guida dettagliate sulla contribuzione.\n\n## 📜 Licenza\n\nAgentDock è rilasciato sotto la [Licenza MIT](../../LICENSE).\n\n## ✨ Crea Possibilità Infinite!\n\nAgentDock fornisce la base per costruire quasi qualsiasi applicazione o automazione basata sull'AI che puoi immaginare. Ti incoraggiamo a esplorare il framework, costruire agenti innovativi e contribuire alla comunità. Costruiamo insieme il futuro dell'interazione AI!\n\n---\n[Torna all'Indice delle Traduzioni](/docs/i18n/README.md)"}}}}