Skip to main content
The Cache API stores values for reuse across requests. Each value belongs to a key and an access scope.

​
Project and user scopes

Project cache operations address data shared within a project. User cache operations address data for the authenticated user. The same key in two scopes does not identify the same cache entry. Credentials and project permissions determine which scope a request can access.

​
Values and expiration

Cache values are strings. Structured data must be serialized before storage and parsed after retrieval. Writes accept ttl in seconds. The default is 300 seconds, the maximum is 86,400 seconds, and 0 disables expiration. For example, a support application might cache a serialized lookup result for reuse. A shared project key and a caller-specific user key have different audiences. Your application decides when cached data is stale and whether to replace or delete it.

​
Batch operations and management

Project cache operations support batches, key listing, statistics, pattern-based deletion, and clearing. Batch reads and writes accept up to 100 entries per request. Cached data supports application behavior but is separate from project source, uploaded files, and execution history. The Projects API manages source and assets; the Execution API manages runs and events.

​
API references