Skip to main content
Glama

MCP DadosBR

.env.example7.5 kB
# ============================================================================= # MCP DadosBR - Environment Configuration Example # ============================================================================= # Copy this file to .env and configure according to your needs # For more information: https://github.com/cristianoaredes/mcp-dadosbr # ----------------------------------------------------------------------------- # Transport Configuration # ----------------------------------------------------------------------------- # Transport mode: stdio (default, for IDE integration) or http (for web deployment) MCP_TRANSPORT=stdio # HTTP server port (only used when MCP_TRANSPORT=http) MCP_HTTP_PORT=3000 # ----------------------------------------------------------------------------- # Cache Configuration # ----------------------------------------------------------------------------- # Maximum number of cache entries # Default: 256 # Valid range: 1-10000 MCP_CACHE_SIZE=256 # Default cache TTL in milliseconds # Default: 60000 (1 minute) # Valid range: 1000-3600000 (1 second to 1 hour) MCP_CACHE_TTL=60000 # Enable background cleanup of expired cache entries # Default: true MCP_CACHE_BACKGROUND_CLEANUP=true # Category-specific cache TTL (optional, overrides MCP_CACHE_TTL) # CNPJ data TTL - Company data is relatively stable # Default: 300000 (5 minutes) MCP_CACHE_TTL_CNPJ=300000 # CEP data TTL - Postal codes are very stable # Default: 600000 (10 minutes) MCP_CACHE_TTL_CEP=600000 # Search results TTL - Search results change frequently # Default: 60000 (1 minute) MCP_CACHE_TTL_SEARCH=60000 # Intelligence report TTL - Derived from searches # Default: 30000 (30 seconds) MCP_CACHE_TTL_INTELLIGENCE=30000 # ----------------------------------------------------------------------------- # API Configuration # ----------------------------------------------------------------------------- # CNPJ API endpoint URL # Default: https://api.opencnpj.org/ # Allowed hosts: api.opencnpj.org, receitaws.com.br, brasilapi.com.br, # publica.cnpj.ws, minhareceita.org, localhost, 127.0.0.1 CNPJ_API_BASE_URL=https://api.opencnpj.org/ # CEP API endpoint URL # Default: https://opencep.com/v1/ # Allowed hosts: opencep.com, viacep.com.br, brasilapi.com.br, # cep.awesomeapi.com.br, localhost, 127.0.0.1 CEP_API_BASE_URL=https://opencep.com/v1/ # Custom authentication headers (optional) # Header name for API authentication API_KEY_HEADER=X-API-Key # API key value (leave empty if not required) API_KEY_VALUE= # ----------------------------------------------------------------------------- # Search Configuration (Required for search tools) # ----------------------------------------------------------------------------- # Tavily API key - Required for cnpj_search and cnpj_intelligence tools # Get your API key at: https://tavily.com # If not set, search functionality will be disabled TAVILY_API_KEY= # ----------------------------------------------------------------------------- # Timeout Configuration # ----------------------------------------------------------------------------- # HTTP request timeout for external API calls (CNPJ, CEP) # Default: 8000ms (8 seconds) # Valid range: 1000-30000ms MCP_API_TIMEOUT=8000 # Total timeout for intelligence search operations # Default: 25000ms (25 seconds) # Valid range: 5000-60000ms MCP_INTELLIGENCE_TIMEOUT=25000 # Timeout for individual search requests # Default: 10000ms (10 seconds) # Valid range: 2000-30000ms MCP_SEARCH_TIMEOUT=10000 # Timeout for request deduplication cleanup # Default: 30000ms (30 seconds) # Valid range: 5000-120000ms MCP_DEDUP_TIMEOUT=30000 # Circuit breaker reset timeout # Default: 30000ms (30 seconds) # Valid range: 10000-300000ms MCP_CIRCUIT_BREAKER_TIMEOUT=30000 # SSE connection timeout (Cloudflare Workers) # Default: 300000ms (5 minutes) # Valid range: 10000-600000ms MCP_SSE_TIMEOUT=300000 # Ping interval for SSE keep-alive # Default: 30000ms (30 seconds) # Valid range: 5000-120000ms # Note: SSE_TIMEOUT should be > 2 * PING_INTERVAL MCP_PING_INTERVAL=30000 # Rate limiter cleanup interval # Default: 60000ms (1 minute) # Valid range: 10000-600000ms MCP_RATE_LIMIT_CLEANUP_INTERVAL=60000 # Rate limit window duration # Default: 60000ms (1 minute) # Valid range: 10000-600000ms MCP_RATE_LIMIT_WINDOW=60000 # ----------------------------------------------------------------------------- # Cloudflare Workers Configuration (Production Deployment) # ----------------------------------------------------------------------------- # API key for Cloudflare Workers endpoints # Use: wrangler secret put MCP_API_KEY # This protects REST endpoints: /cnpj/*, /cep/*, /search, /intelligence, /thinking # Note: /mcp and /sse endpoints remain unauthenticated for AI assistant compatibility # MCP_API_KEY=your-secret-key-here # Disable rate limiting (not recommended for production) # Default: false # Set to true to disable rate limiting entirely MCP_DISABLE_RATE_LIMIT=false # ----------------------------------------------------------------------------- # Logging and Debugging # ----------------------------------------------------------------------------- # Node environment: development | production # Default: development # Production mode disables debug logs and masks sensitive information NODE_ENV=production # Log level: debug | info | warn | error # Default: info (in production), debug (in development) # - debug: All logs including internal debugging information # - info: General informational messages # - warn: Warning conditions # - error: Error conditions only LOG_LEVEL=info # Disable sequential thinking logs # Default: false # Set to true to disable verbose thinking process logs DISABLE_THOUGHT_LOGGING=false # ----------------------------------------------------------------------------- # Rate Limiting (Local Mode - stdio/http) # ----------------------------------------------------------------------------- # Rate limit time window in milliseconds # Default: 60000 (1 minute) RATE_LIMIT_WINDOW_MS=60000 # Maximum requests per window per client # Default: 30 RATE_LIMIT_MAX_REQUESTS=30 # ----------------------------------------------------------------------------- # Advanced Configuration (Use with caution) # ----------------------------------------------------------------------------- # Maximum concurrent searches in intelligence operations # Default: 3 # Valid range: 1-10 # MAX_CONCURRENT_SEARCHES=3 # Maximum results to cache for search queries # Default: 20 # MAX_RESULTS_CACHED=20 # ----------------------------------------------------------------------------- # Scenario-Based Configuration Examples # ----------------------------------------------------------------------------- # Example 1: Development/Testing (Fast, short TTLs) # NODE_ENV=development # LOG_LEVEL=debug # MCP_CACHE_TTL=30000 # MCP_API_TIMEOUT=5000 # Example 2: Production (Secure, long TTLs, masked logs) # NODE_ENV=production # LOG_LEVEL=info # MCP_CACHE_TTL_CNPJ=600000 # MCP_CACHE_TTL_CEP=1200000 # MCP_API_TIMEOUT=10000 # Example 3: High-Performance (Aggressive caching) # MCP_CACHE_SIZE=1024 # MCP_CACHE_TTL_CNPJ=900000 # MCP_CACHE_TTL_CEP=1800000 # Example 4: Slow Network (Longer timeouts) # MCP_API_TIMEOUT=15000 # MCP_SEARCH_TIMEOUT=20000 # MCP_INTELLIGENCE_TIMEOUT=45000 # Example 5: Cloudflare Workers Deployment # NODE_ENV=production # LOG_LEVEL=warn # MCP_DISABLE_RATE_LIMIT=false # Use wrangler secrets for MCP_API_KEY and TAVILY_API_KEY

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/cristianoaredes/mcp-dadosbr'

If you have feedback or need assistance with the MCP directory API, please join our Discord server