mastra-ai/mastra · error · MastraError

EXPERIMENT_NO_ITEMS

EXPERIMENT_NO_ITEMS

Error message

No items in dataset ${datasetId} at version ${datasetVersion}

What it means

After resolving the dataset, runExperiment fetches the items for the dataset's version via getItemsByVersion; if the version has zero items the experiment would be meaningless, so a MastraError with id EXPERIMENT_NO_ITEMS is thrown. This guards against silently 'running' an experiment over an empty item set.

Source

Thrown at packages/core/src/datasets/experiment/index.ts:275

      datasetRecord = await datasetsStore.getDatasetById({ id: datasetId, filters: config.filters });
      if (!datasetRecord) {
        throw new MastraError({
          id: 'DATASET_NOT_FOUND',
          text: `Dataset not found: ${datasetId}`,
          domain: 'STORAGE',
          category: 'USER',
        });
      }

      datasetVersion = version ?? datasetRecord.version;
      const versionItems = await datasetsStore.getItemsByVersion({
        datasetId,
        version: datasetVersion,
      });

      if (versionItems.length === 0) {
        throw new MastraError({
          id: 'EXPERIMENT_NO_ITEMS',
          text: `No items in dataset ${datasetId} at version ${datasetVersion}`,
          domain: 'STORAGE',
          category: 'USER',
        });
      }

      items = versionItems.map(v => ({
        id: v.id,
        datasetVersion: v.datasetVersion,
        input: v.input,
        groundTruth: v.groundTruth,
        expectedTrajectory: v.expectedTrajectory as TrajectoryExpectation | undefined,
        requestContext: v.requestContext,
        metadata: v.metadata,
        toolMocks: v.toolMocks,
        unmockedToolPolicy: v.unmockedToolPolicy,
        scorerIds: v.scorerIds,

View on GitHub (pinned to 75dd419e61)

Solutions

Add items to the dataset version via the datasets store (addItems/upload) before running
Target the dataset version that actually contains items
Check getItemsByVersion output to confirm the item count for that version
If items were purged, restore/re-upload them and re-run

Example fix

// before
await runExperiment({ datasetId: 'ds_123', ... }); // version 2 empty
// after
const items = await datasetsStore.getItemsByVersion({ datasetId: 'ds_123', version: 2 });
if (items.length === 0) await datasetsStore.addItems({ datasetId: 'ds_123', items });
await runExperiment({ datasetId: 'ds_123', ... });

Defensive patterns

Strategy: validation

Validate before calling

const items = await datasetsStore.getItemsByVersion({ datasetId, version });
if (items.length === 0) throw new Error(`Dataset ${datasetId} v${version} has no items`);

Try / catch

try {
  await runExperiment(config);
} catch (err) {
  if (err instanceof MastraError && err.id === 'EXPERIMENT_NO_ITEMS') {
    // add items to the dataset version, then retry
  } else throw err;
}

Prevention

Upload items immediately after dataset creation and verify the count
Check item counts per version before scheduling experiment runs
Confirm which dataset version the run will target
Alert on datasets whose latest version has zero items

When it happens

Trigger: Running an experiment against a dataset version that contains no items — e.g. a newly created dataset whose items were never added, or a version where all items were deleted.

Common situations: Dataset created but item upload failed or was skipped; pointing at version N before items were pushed; retention job purged items; using the latest version identifier when items live on an earlier version.

Related errors

AI-assisted analysis of mastra-ai/mastra@75dd419e61 (2026-08-30). Data as JSON: /api/errors/d8b290522707f4e9. Report an issue: GitHub.