YouTube Data API client documentation

Badacz odbiorców

Private, non-commercial audience-research tool for one owner.

Current documented state

What this client is

Badacz odbiorców is a private, non-commercial local tool used by one owner. It is not sold, distributed to customers, offered as a public dashboard, or used to contact YouTube users. This public page documents the API client and its limits; it does not provide access to the tool.

The intended use is analytics and reporting: documenting how a research topic can be translated into traceable source-discovery queries and, only if the relevant YouTube terms are accepted and the client is permitted to do so, producing independently generated analytical observations. Any independently generated metric or category will be clearly labelled as independently generated and not as a metric supplied by YouTube.

Current capabilities and boundaries

Implemented

Synthetic demonstration

A local, Synthetic demonstration validates a zero-cost scenario and blocks a requested comment operation when the YouTube terms state is unresolved. It makes no network request and is not a market result.

Implemented, narrow pilot

Minimal metadata search

A one-time technical pilot used search.list for up to five public video records. It did not paginate, create derived metrics, retrieve comments, or establish a market finding.

Blocked

Full live source discovery

The planned 12 Polish plus 12 English queries are not run. The local code stops before a live request while access remains unresolved.

Not implemented

Comment collection

commentThreads.list and comments.list, reply retrieval, resumption, comment deduplication and raw-data deletion or refresh controls are not implemented and are not currently retrieved.

How YouTube API data would be handled if permitted

The future workflow is: a research brief → official YouTube Data API source discovery → local storage of the permitted minimum → separately labelled analytical output → a traceable local report. It does not use scraping, does not automate engagement, does not publish on YouTube, does not profile authors and does not train a model on comments.

Until the relevant permission and implementation exist, the future workflow remains disabled. The complete distinction between current and planned processing appears in the data map.

Public documents