Zebu Data - Customer Reviews

The five tools map to fairly distinct resources and actions: aborting a run, inspecting a run, retrieving dataset rows, retrieving a key-value record, and running the Yelp scraper. There is slight potential overlap between the scraper tool returning output and get-dataset-items, but the descriptions make the intended separation reasonably clear.

Four tools follow a readable kebab-case verb-noun pattern: abort-actor-run, get-actor-run, get-dataset-items, get-key-value-store-record. However, delicious_zebu--yelp-reviews-scraper uses an actor-derived name with a double hyphen and no leading action verb, making the set mixed but still understandable.

Five tools is well-scoped for an actor-based customer reviews workflow. The set covers running the scraper, checking or aborting the run, and reading the resulting dataset or key-value store without bloat.

The surface covers the core end-to-end workflow: launch the Yelp reviews scraper, monitor or abort the run, and retrieve output rows or stored records. Minor gaps exist around discovery/listing operations, such as listing runs or datasets, but agents can work around these using known identifiers.