mcp_prompt

pulseengine_mcp_macros

Attribute Macro mcp_prompt 

Source 
#[mcp_prompt]
Expand description

Automatically generates MCP resource definitions from Rust functions.

This macro transforms regular Rust functions into MCP resources with automatic URI template parsing, parameter extraction, and content type handling.

§Basic Usage

ⓘ
use pulseengine_mcp_macros::mcp_resource;

#[mcp_resource(uri_template = "file://{path}")]
async fn read_file(&self, path: String) -> Result<String, std::io::Error> {
    tokio::fs::read_to_string(&path).await
}

§With Custom Configuration

ⓘ
#[mcp_resource(
    uri_template = "db://{database}/{table}",
    name = "database_table",
    description = "Read data from a database table",
    mime_type = "application/json"
)]
async fn read_table(&self, database: String, table: String) -> Result<serde_json::Value, Error> {
    // Implementation
}

§Parameters

  • uri_template: Required URI template with parameters in {param} format
  • name: Optional custom resource name (defaults to function name)
  • description: Optional custom description (defaults to doc comments)
  • mime_type: Optional MIME type (defaults to “text/plain”)

§Features

  • URI Template Parsing: Automatic extraction of parameters from URI templates
  • Type Safety: Compile-time validation of parameter types
  • Auto-Documentation: Uses function doc comments as resource descriptions
  • Content Type Detection: Automatic MIME type handling
  • Error Handling: Converts function errors to MCP protocol errors

§References

This macro transforms regular Rust functions into MCP prompts with automatic argument validation and prompt message generation.

§Basic Usage

ⓘ
use pulseengine_mcp_macros::mcp_prompt;

#[mcp_prompt(name = "code_review")]
async fn generate_code_review(&self, code: String, language: String) -> Result<PromptMessage, Error> {
    // Generate prompt for code review
}

§With Custom Configuration

ⓘ
#[mcp_prompt(
    name = "sql_query_helper",
    description = "Generate SQL queries based on natural language",
    arguments = ["description", "table_schema", "output_format"]
)]
async fn sql_helper(&self, description: String, table_schema: String, output_format: String) -> Result<PromptMessage, Error> {
    // Implementation
}

§Parameters

  • name: Optional custom prompt name (defaults to function name)
  • description: Optional custom description (defaults to doc comments)
  • arguments: Optional array of argument names for documentation

§Features

  • Argument Validation: Automatic validation of prompt arguments
  • Type Safety: Compile-time validation of parameter types
  • Auto-Documentation: Uses function doc comments as prompt descriptions
  • Error Handling: Converts function errors to MCP protocol errors
  • Schema Generation: Automatic argument schema generation

§References