🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
/**
* MCP Error Classes
*
* Following Sentry MCP pattern:
* - UserInputError: 4xx-like errors that user can fix (invalid input, not found, etc.)
* - ApiServerError: 5xx-like errors that should be logged to monitoring
*/
/**
* Error that indicates user input is invalid or incorrect.
* These errors should be formatted for user display and not logged as server errors.
*
* Examples:
* - Invalid prompt name
* - Prompt not found
* - Invalid version number
* - Missing required fields
*/
export class UserInputError extends Error {
constructor(message: string) {
super(message);
this.name = "UserInputError";
// Maintains proper stack trace for where our error was thrown (only available on V8)
if (Error.captureStackTrace) {
Error.captureStackTrace(this, UserInputError);
}
}
}
/**
* Error that indicates an internal server problem.
* These errors should be logged to monitoring systems.
*
* Examples:
* - Database connection failure
* - Redis unavailable
* - Unexpected null value
* - Third-party API failure
*/
export class ApiServerError extends Error {
constructor(message: string, options?: ErrorOptions) {
super(message, options);
this.name = "ApiServerError";
// Maintains proper stack trace for where our error was thrown (only available on V8)
if (Error.captureStackTrace) {
Error.captureStackTrace(this, ApiServerError);
}
}
}
/**
* Type guard to check if an error is a UserInputError
*/
export function isUserInputError(error: unknown): error is UserInputError {
return error instanceof UserInputError;
}
/**
* Type guard to check if an error is an ApiServerError
*/
export function isApiServerError(error: unknown): error is ApiServerError {
return error instanceof ApiServerError;
}