eyaltoledano/claude-task-master · error · MCPError

Generated object does not match schema: ${validationError.me

Error message

Generated object does not match schema: ${validationError.message}. Generated: ${JSON.stringify(parsedObject, null, 2)}

What it means

After parsing the model's JSON, doGenerateObject validates it against the provided schema; if validation fails it throws an MCPError containing the validator's message plus the full pretty-printed generated object. This means the model produced valid JSON that doesn't conform to the requested structure (missing fields, wrong types, extra/invalid values).

Source

Thrown at mcp-server/src/custom-sdk/language-model.js:173

			// Validate against schema
			try {
				const validatedObject = schema.parse(parsedObject);

				return {
					object: validatedObject,
					finishReason: result.finishReason || 'stop',
					usage: {
						promptTokens: result.usage?.inputTokens || 0,
						completionTokens: result.usage?.outputTokens || 0,
						totalTokens:
							(result.usage?.inputTokens || 0) +
							(result.usage?.outputTokens || 0)
					},
					rawResponse: response,
					warnings: result.warnings
				};
			} catch (validationError) {
				throw new MCPError(
					`Generated object does not match schema: ${validationError.message}. Generated: ${JSON.stringify(parsedObject, null, 2)}`
				);
			}
		} catch (error) {
			throw mapMCPError(error);
		}
	}

	/**
	 * Stream text generation using MCP session sampling
	 * Note: MCP may not support native streaming, so this may simulate streaming
	 * @param {object} options - Generation options
	 * @returns {AsyncIterable} Stream of generation chunks
	 */
	async doStream(options) {
		try {
			// For now, simulate streaming by chunking the complete response
			// TODO: Implement native streaming if MCP supports it

View on GitHub (pinned to c0c98d367c)

Solutions

Make the schema lenient where possible (optional fields, defaults, unions) and re-prompt with explicit field requirements and an example output
Retry the generation — schema-adherence failures are often stochastic; add 'Respond ONLY with JSON matching this exact schema' instructions
Post-process/repair the object (fill defaults, coerce types) before validation, or use a repair/re-ask loop

Example fix

// before
const schema = z.object({ count: z.number() });
// after
const schema = z.object({ count: z.coerce.number().default(0) });

Defensive patterns

Strategy: try-catch

Validate before calling

// Pre-validate expected fields if you control the prompt/output contract
if (!jsonText.includes('"requiredField"')) console.warn('Response may be missing required fields');

Try / catch

try { obj = await model.doGenerateObject({ schema, prompt }); } catch (e) {
  if (e.message.startsWith('Generated object does not match schema')) {
    // retry with a stricter prompt or relax the schema
  } else throw e;
}

Prevention

Include an explicit JSON example matching the schema in the prompt
Keep schemas simple: avoid strict enums/formats the model may violate; mark non-essential fields optional
Implement one retry-with-feedback loop: feed the validation error back to the model on failure

When it happens

Trigger: Model returns JSON whose shape fails schema validation — missing required properties, wrong types (string vs number), or values violating constraints — when calling doGenerateObject/generateObject through the MCP language model.

Common situations: Weak models hallucinating fields or omitting required ones; overly strict schemas (enums, formats) the model can't satisfy; schema/prompt mismatch after refactoring.

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

AI-assisted analysis of eyaltoledano/claude-task-master@c0c98d367c (2026-08-29). Data as JSON: /api/errors/2cf616d8040dd58d. Report an issue: GitHub.