What causes "AI response did not include a valid subtasks array." in eyaltoledano/claude-task-master?

Retry the expand command; transient provider issues often produce empty results.

AI response did not include a valid subtasks array.

Error message

AI response did not include a valid subtasks array.

What it means

expandTask() asks the AI service to generate subtasks via generateObjectService and expects the response's mainResult to contain a subtasks array (structured generateObject output). If mainResult is missing or subtasks is not an array, it cannot append anything and throws before mutating the task.

Source

Thrown at scripts/modules/task-manager/expand-task.js:332

			const role = useResearch ? 'research' : 'main';

			// Call generateObjectService with the determined prompts and telemetry params
			aiServiceResponse = await generateObjectService({
				prompt: promptContent,
				systemPrompt: systemPrompt,
				role,
				session,
				projectRoot,
				schema: COMMAND_SCHEMAS['expand-task'],
				objectName: 'subtasks',
				commandName: 'expand-task',
				outputType: outputFormat
			});

			// With generateObject, we expect structured data – verify it before use
			const mainResult = aiServiceResponse?.mainResult;
			if (!mainResult || !Array.isArray(mainResult.subtasks)) {
				throw new Error('AI response did not include a valid subtasks array.');
			}
			generatedSubtasks = mainResult.subtasks.map((subtask) => ({
				...subtask,
				dependencies: subtask.dependencies ?? [],
				status: subtask.status ?? 'pending',
				testStrategy: subtask.testStrategy ?? null
			}));
			logger.info(`Received ${generatedSubtasks.length} subtasks from AI.`);
		} catch (error) {
			if (loadingIndicator) stopLoadingIndicator(loadingIndicator);
			logger.error(
				`Error during AI call or parsing for task ${taskId}: ${error.message}`, // Added task ID context
				'error'
			);
			throw error;
		} finally {
			if (loadingIndicator) stopLoadingIndicator(loadingIndicator);
		}

View on GitHub (pinned to c0c98d367c)

Solutions

Retry the expand command; transient provider issues often produce empty results.
Verify your model configuration (task-master models) and use a model that reliably follows JSON schemas.
Check that the zod schema passed to generateObjectService requires a subtasks array so failures are retried upstream, and confirm ai-services-client version matches task-master's expectation of { mainResult: { subtasks: [...] } }.

Example fix

// before (service layer)
return { mainResult: parsed };   // parsed = { subtask: {...} }
// after
return { mainResult: { subtasks: Array.isArray(parsed.subtasks) ? parsed.subtasks : [parsed.subtask] } };

Defensive patterns

Strategy: try-catch

Type guard

function hasSubtasksArray(res) {
  return !!res?.mainResult && Array.isArray(res.mainResult.subtasks);
}

Try / catch

try {
  await expandTask(taskId, context);
} catch (err) {
  if (err.message.includes('valid subtasks array')) {
    console.error('AI returned no subtasks array — check model config and retry, or add subtasks manually.');
  } else throw err;
}

Prevention

Use a schema-compliant model for subtask generation (task-master models).
Ensure the generation schema strictly requires subtasks: z.array().
Retry once automatically; provider hiccups often cause empty results.

When it happens

Trigger: generateObjectService returns undefined/absent mainResult, or mainResult lacks a subtasks array — e.g. the model returned a single object instead of an array, schema validation silently degraded, or the service returned an error envelope.

Common situations: Weak/non-schema-following models returning { subtask: {...} } or { subtasks: {...} } (object not array); provider errors from bad API keys or rate limits surfaced as empty results; version drift between task-master and the AI services module changing the response shape.

Related errors

AI-assisted analysis of eyaltoledano/claude-task-master@c0c98d367c (2026-08-29). Data as JSON: /api/errors/32e1a8fdb4fdae83. Report an issue: GitHub.