yikart/AiToEarn · error · McpError

InvalidParams

InvalidParams

Error message

Invalid parameters: ${validation.error.message}

What it means

Before executing a tool, the handler validates request.params.arguments against the tool's Zod parameters schema. If safeParse fails, it throws McpError InvalidParams (JSON-RPC -32602), because the arguments do not satisfy the tool's declared input contract.

Source

Thrown at project/aitoearn-backend/libs/nest-mcp/src/services/handlers/mcp-tools.handler.ts:134

          this.mcpModuleId,
          request.params.name,
        )

        if (!toolInfo) {
          throw new McpError(
            ErrorCode.MethodNotFound,
            `Unknown tool: ${request.params.name}`,
          )
        }

        try {
          // Validate input parameters against the tool's schema
          if (toolInfo.metadata.parameters) {
            const validation = toolInfo.metadata.parameters.safeParse(
              request.params.arguments || {},
            )
            if (!validation.success) {
              throw new McpError(
                ErrorCode.InvalidParams,
                `Invalid parameters: ${validation.error.message}`,
              )
            }
            // Use validated arguments to ensure defaults and transformations are applied
            request.params.arguments = validation.data
          }

          const contextId = ContextIdFactory.getByRequest(httpRequest)
          this.moduleRef.registerRequestByContextId(httpRequest, contextId)

          const toolInstance = await this.moduleRef.resolve(
            toolInfo.providerClass,
            contextId,
            { strict: false },
          )

          const context = this.createContext(mcpServer, request)

View on GitHub (pinned to d3aa8bea5b)

Solutions

1. Read the validation.error.message for the exact failing field and fix the arguments
2. Validate arguments client-side against the tool's inputSchema before calling
3. Refetch tools/list to get the current inputSchema after any server update
4. Coerce types (e.g. Number(id)) for arguments supplied as strings

Example fix

// before
await client.callTool({ name: 'createPost', arguments: { title: 'x' } });
// after
const args = parameters.parse({ title: 'x' }); // throws with the same message early
await client.callTool({ name: 'createPost', arguments: args });

import { z } from 'zod';
const parsed = inputSchema.safeParse(args);
if (!parsed.success) {
  throw new Error(`Invalid arguments: ${parsed.error.message}`);
}
await client.callTool({ name, arguments: parsed.data });

function isValidArgs(args: unknown, schema: z.ZodTypeAny): args is z.infer<typeof schema> {
  return schema.safeParse(args).success;
}

try {
  await client.callTool({ name, arguments: args });
} catch (e) {
  if (e.code === ErrorCode.InvalidParams) {
    console.error('Fix arguments per tool inputSchema:', e.message);
  }
  throw e;
}

When it happens

Trigger: Missing required arguments, wrong argument types (string vs number), arguments passed under wrong keys, or arguments omitted entirely when required fields exist ({} is validated).

Common situations: Client built arguments by hand instead of from the tool's inputSchema; schema changed server-side while the client cached an older version; nested object/array arguments malformed; numbers sent as strings from LLM-generated arguments.

Understand the failure class

Background: Schema validation failed / invalid input schema: payload rejected because its shape doesn't match the expected schema — this error's family across 28 libraries.

Related errors

• InternalError
• MethodNotFound
• PlatformNotSupported
• Zod validation error (issues + input)
• ResponseCode.ConfigEditorValidationFailed

AI-assisted analysis of yikart/AiToEarn@d3aa8bea5b (2026-08-31). Data as JSON: /api/errors/aa90de3537fd0085. Report an issue: GitHub.