Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_HOST | No | HTTP host | 0.0.0.0 |
| MCP_PORT | No | HTTP port | 8080 |
| OTEL_ENABLED | No | Enable OpenTelemetry | true |
| MCP_LOG_LEVEL | No | Log level | INFO |
| MCP_TRANSPORT | No | Transport type (stdio or http) | stdio |
| STANDARDS_PATH | No | Path to standards docs | docs/standards |
| MCP_SERVER_NAME | No | Server name | standards-mcp-server |
| OTEL_EXPORTER_OTLP_ENDPOINT | No | OTLP endpoint | |
| REFERENCE_IMPLEMENTATIONS_PATH | No | Path to implementations | reference-implementations |
Capabilities
Features and capabilities supported by this server
| Capability | Details |
|---|---|
| tasks | {
"list": {},
"cancel": {},
"requests": {
"tools": {
"call": {}
},
"prompts": {
"get": {}
},
"resources": {
"read": {}
}
}
} |
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| validate_azure_resource_name | Validate an Azure resource name against naming conventions. Checks if the given resource name follows the Darwin platform naming conventions for Azure resources. Args: resource_name: The resource name to validate resource_type: Type of Azure resource (e.g., "resource_group", "storage_account") ctx: MCP context for logging Returns: ValidationResult with any issues found. Example: >>> result = await validate_azure_resource_name( ... "rg-myapp-eus-dev", ... "resource_group" ... ) >>> print(result["valid"]) True |
| validate_mcp_tool_definition | Validate an MCP tool definition against standards. Checks if the tool follows Darwin platform MCP tool design standards. Args: tool_name: Name of the tool tool_description: Tool docstring/description parameters: Tool parameters schema (JSON Schema format) ctx: MCP context Returns: ValidationResult with any issues found. |
| validate_agent_card | Validate an Agent Card against A2A protocol standards. Checks if the agent card follows the Darwin platform A2A standards and contains all required fields. Args: agent_card: The agent card JSON to validate ctx: MCP context Returns: ValidationResult with any issues found. |
| search_standards | Search standards documents for relevant content. Performs a full-text search across all standards documents and returns relevant excerpts with context. Args: query: Search query (keywords or phrases) category: Optional category filter (agents, mcp, infrastructure) max_results: Maximum number of results to return (default: 10) ctx: MCP context Returns: Dictionary with search results and metadata. Example: >>> results = await search_standards("authentication") >>> for r in results["results"]: ... print(f"{r['title']}: {r['excerpt']}") |
| list_standards_categories | List all available standards categories. Returns a list of all top-level categories in the standards documentation with descriptions and document counts. Args: ctx: MCP context Returns: Dictionary with category information. |
| get_standards_summary | Get a summary of all standards in a category. Returns titles and descriptions of all documents in the specified category for quick reference. Args: category: Category name (agents, mcp, infrastructure) ctx: MCP context Returns: Dictionary with document summaries. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| agent_design_standards | Agent design principles and 5-layer architecture. This document defines the core design principles for building AI agents in the Darwin platform. Topics covered: - 5-layer agent architecture - Agent lifecycle - Error handling patterns - Testing requirements Cache: Long-lived (24 hours) |
| agent_communication_standards | Agent-to-Agent (A2A) protocol standards. Defines how agents communicate with each other using the A2A protocol based on Google's specification. Topics covered: - A2A protocol overview - Agent Cards - Task lifecycle - Message formats Cache: Long-lived (24 hours) |
| agent_tools_standards | MCP tool integration standards for agents. Defines how agents should integrate with and use MCP tools. Topics covered: - Tool design principles - Context usage - Error handling - Testing patterns Cache: Long-lived (24 hours) |
| agent_memory_standards | Agent memory architecture using Neo4j. Defines the graph-based memory system for agents. Topics covered: - Memory types (episodic, semantic, procedural) - Neo4j schema - Memory operations - Retention policies Cache: Long-lived (24 hours) |
| agent_observability_standards | Agent observability standards with OpenTelemetry. Defines observability requirements for agents using OpenTelemetry as the instrumentation foundation. Topics covered: - OpenTelemetry SDK setup - Trace propagation - LLM-specific metrics - Langfuse integration Cache: Long-lived (24 hours) |
| agent_security_standards | Agent security requirements. Defines security requirements for agent authentication, authorization, and secrets management. Topics covered: - Authentication methods - Role-based access control - Secrets management - Audit logging Cache: Long-lived (24 hours) |
| mcp_server_design_standards | MCP server architecture standards. Defines how to build MCP servers for the Darwin platform. Topics covered: - Server architecture - Transport configuration - Capability advertisement - Health checks Cache: Long-lived (24 hours) |
| mcp_tool_design_standards | MCP tool definition standards. Defines how to design and implement MCP tools. Topics covered: - Tool design principles - Input/output schemas - Tool categories - Documentation requirements Cache: Long-lived (24 hours) |
| mcp_resource_design_standards | MCP resource exposure standards. Defines how to design and implement MCP resources. Topics covered: - Resource types - URI schemes - Caching strategies - Subscriptions Cache: Long-lived (24 hours) |
| mcp_security_standards | MCP authentication and authorization standards. Defines security requirements for MCP servers. Topics covered: - Transport security - Authentication methods - Authorization patterns - Audit logging Cache: Long-lived (24 hours) |
| mcp_error_handling_standards | MCP error handling patterns. Defines error handling standards for MCP servers. Topics covered: - Error categories - Error response structure - Retry logic - Error logging Cache: Long-lived (24 hours) |
| mcp_template_python | Python MCP server template - main server file. A complete, standards-compliant Python MCP server template using FastMCP framework. Use this as a starting point for new Python MCP servers. |
| mcp_template_python_config | Python MCP server template - configuration module. Pydantic-based configuration with environment variable support. |
| mcp_template_python_dockerfile | Python MCP server template - Dockerfile. Multi-stage Docker build for production deployment. |
| mcp_template_python_pyproject | Python MCP server template - pyproject.toml. Project configuration with dependencies and build settings. |
| mcp_template_dotnet | C#/.NET MCP server template - main program file. A standards-compliant .NET MCP server template. |
| mcp_impl_standards | Standards MCP server implementation - main server file. Reference implementation of an MCP server providing standards documentation and validation tools. |
| mcp_impl_standards_validation | Standards MCP server - validation tools implementation. Example of MCP tool implementation for validation. |
| mcp_impl_standards_search | Standards MCP server - search tools implementation. Example of MCP tool implementation for searching. |
| mcp_impl_standards_resources | Standards MCP server - resources implementation. Example of MCP resource implementation. |
| mcp_impl_healthcare | Healthcare MCP server implementation - README. Domain-specific MCP server for healthcare/pharmacy operations. |
| agent_impl_langchain | LangChain agent implementation - README. Reference implementation of an agent using LangChain framework. |
| agent_impl_langgraph | LangGraph agent implementation - README. Reference implementation of an agent using LangGraph framework. |
| agent_impl_dotnet | MS Agent Framework implementation - README. Reference implementation using Microsoft Agent Framework. |
| a2a_impl_python | Python A2A implementation - README. Reference implementation of A2A protocol in Python. |
| a2a_impl_dotnet | C#/.NET A2A implementation - README. Reference implementation of A2A protocol in .NET. |