eBay Sold Listings Price Decision Pack

Post-process datasets from eBay Sold Listings Search into deterministic decision-ready rows. This Actor does not run the upstream Actor by itself; it reads an explicitly provided datasetId/defaultDatasetId or equivalent direct records.

What It Does

It validates the upstream dataset contract, normalizes required fields, removes duplicates, preserves source attribution, applies explicit rule scoring, and emits a new output contract for downstream workflows.

Who It Is For

Operators who already run the upstream Actor and need a bounded post-processing layer with repeatable validation, scoring, and integration-ready output.

Input

Use datasetId or defaultDatasetId for an existing Apify dataset, or provide records directly. Required upstream fields: title, soldPrice, currency, soldAt.

Output

Each row includes sourceActorId, sourceDatasetId, sourceRecordIndex, normalizedKey, score, decision, ScoreExplanation, and sourceAttribution.

How It Works

The F1 pattern is schema-first: resolve dataset or direct records, validate required upstream fields, normalize, deduplicate, score with deterministic thresholds, attach source attribution, and emit only valid post-processed rows.

Recommended action derivation is explicit: score 80 or higher means prioritize, score 65-79 means review, and lower scores mean monitor. Score contributions are required-field completeness plus observed numeric signal strength; every emitted row includes ScoreExplanation.

Example

{ "datasetId": "abc123", "maxRecords": 5000 }

Pricing

PAY_PER_EVENT event post_processed_record_emitted is charged once per valid post-processed row emitted. Invalid records, duplicates, empty datasets, and rows skipped by charge limits are not billable.

Limitations

This is not a wrapper pass-through and it does not claim to acquire upstream data. Upstream acquisition cost is separate from post-processor cost. If the upstream schema fingerprint changes incompatibly, the Actor fails closed with a schema mismatch report.

Integration

Use it standalone after an upstream run or as an integration target in a workflow. Direct upstream execution may be added later only when it can stay within LIMITED_PERMISSIONS and explicit cost accounting.