n8n-io/n8n · error · Error
No examples found in dataset "${datasetName}"
Error message
No examples found in dataset "${datasetName}" What it means
No filters were applied but the loaded dataset yielded zero examples and `maxExamples` is set. This is the 'dataset is empty' twin of the filtered-no-match case: the harness cannot run because there is nothing to evaluate. The dataset name is included in the message.
Source
Thrown at packages/@n8n/ai-workflow-builder.ee/evaluations/harness/runner.ts:396
}
if (filters && matches.length === 0) {
const filterSummary = [
filters.notionId ? `id:${filters.notionId}` : undefined,
filters.technique ? `technique:${filters.technique}` : undefined,
filters.doSearch ? `do:${filters.doSearch}` : undefined,
filters.dontSearch ? `dont:${filters.dontSearch}` : undefined,
]
.filter((v): v is string => v !== undefined)
.join(', ');
throw new Error(
`No examples matched filters (${filterSummary}) in dataset "${datasetName}" (scanned ${scanned})`,
);
}
if (!filters && maxExamples && matches.length === 0) {
throw new Error(`No examples found in dataset "${datasetName}"`);
}
return matches;
}
async function resolveLangsmithData(params: {
dataset: string;
langsmithOptions: LangsmithRunConfig['langsmithOptions'];
lsClient: {
readDataset: (args: { datasetName: string }) => Promise<{ id: string }>;
listExamples: (args: { datasetId: string; limit?: number }) => AsyncIterable<Example>;
};
logger: EvalLogger;
}): Promise<string | Example[]> {
const { dataset, langsmithOptions, lsClient, logger } = params;
const datasetName = dataset;
const maxExamples = langsmithOptions.maxExamples;View on GitHub (pinned to 5ac6606e81)
Solutions
- Confirm the dataset path/name is correct and that the file is present and non-empty.
- Re-generate the dataset from its source and re-run.
- Check that `maxExamples` is unset or `> 0`.
Example fix
// before run --dataset empty-dataset --max-examples 10 // after run --dataset real-dataset --max-examples 10
Defensive patterns
Strategy: validation
Validate before calling
function assertDatasetUsable(examples: unknown[], maxExamples?: number): void {
if (examples.length === 0) throw new Error('dataset is empty; cannot run evaluation');
if (typeof maxExamples === 'number' && maxExamples <= 0) throw new Error('maxExamples must be > 0');
}
assertDatasetUsable(loadedExamples, args.maxExamples); Prevention
- Validate the dataset is non-empty before invoking the runner, with a clearer message than the runner's.
- Wire dataset generation into CI so an empty dataset fails the build, not the eval run.
- Default `maxExamples` to `undefined` (run all) unless you explicitly want a slice.
When it happens
Trigger: Dataset file/object resolves to an empty array, or a directory dataset has no valid entries. The `!filters && maxExamples && matches.length === 0` branch.
Common situations: Pointing `--dataset` at the wrong (empty) file; dataset generation script produced no rows due to an upstream error; a gitignored/missing dataset path; `maxExamples=0` passed explicitly.
Related errors
- LangSmith mode requires `--dataset` and does not support `--
- No valid checks after filtering. Requested: ${options.checks
- No examples matched filters (${filterSummary}) in dataset "$
- Model ID is required
- Invalid delegate sub-agent tool name "${name}": must start w
AI-assisted analysis of n8n-io/n8n@5ac6606e81 (2026-08-12).
Data as JSON: /api/errors/15e66d5f57635b9a.
Report an issue: GitHub.