Read and write records in an Apify key-value store, including a run's default store, INPUT and OUTPUT

Documented steps

1. Every Actor run is assigned its own key-value store at creation; its ID is the defaultKeyValueStoreId field on the Run object, and it can also be addressed as 'default' from inside the run.
2. The run's input lives under the reserved key INPUT in that store. Inside Actor code read it via Actor.getInput() (JS) or Actor.get_input() (Python). Locally, the CLI reads INPUT.json from the local storage directory.
3. Write a record: PUT https://api.apify.com/v2/key-value-stores/{storeId}/records/{recordKey} with the raw value as the request body and 'Content-Type' set to the value's own MIME type. The stored content type is what the API will serve back later.
4. Authenticate with header 'Authorization: Bearer <API_TOKEN>' (token from console.apify.com Integrations page). The docs mark the '?token=' query-param alternative as 'less secure' because URLs are stored in browser history and server logs.
5. Optionally set 'Content-Encoding' (br, gzip, deflate, identity) on the PUT to store the record compressed; the encoding is preserved on read.
6. Read a record: GET https://api.apify.com/v2/key-value-stores/{storeId}/records/{recordKey}. The response Content-Type is whatever was stored with the record. Add '?attachment=true' to force a download rather than inline rendering.
7. Read an Actor's result after a run finishes: GET /v2/key-value-stores/{defaultKeyValueStoreId}/records/OUTPUT. Inside an Actor, write it with Actor.setValue('OUTPUT', value).
8. Enumerate a store: GET https://api.apify.com/v2/key-value-stores/{storeId}/keys with limit (1-1000, default 1000), exclusiveStartKey as the pagination cursor, and prefix to filter. The response carries count, limit, isTruncated and nextExclusiveStartKey.
9. Docs: https://docs.apify.com/platform/storage/key-value-store and https://docs.apify.com/api/v2/key-value-store-record-put