Collaboration specification¶
Defines the high-level collaboration. The specification defines which analysis runners are invited, and for each analysis runner, which data and templates they can access and run. Any templates or data offerings that are listed here must be registered before they’re included in the collaboration specification.
The owner submits this specification by calling INITIALIZE.
Schema:
api_versionThe version of the Collaboration API used. Must be
2.0.0.spec_typeSpecification type identifier. Must be
collaboration.name: collaboration_nameUser-friendly name for this collaboration. Must be unique in the creator’s account and follow Snowflake identifier rules (maximum 75 characters).
version(Optional)A version identifier for this collaboration (maximum 20 characters). Must follow Snowflake identifier rules. A good format to use is YYYY_MM_DD_V#. For example:
2025_10_22_V1.description: collaboration_description(Optional)A human-readable description of the collaboration (maximum 1,000 characters), for collaborators to read.
owner: owner_aliasAlias of the collaboration owner, as defined in
collaborator_identifier_aliases.collaborator_identifier_aliasesA mapping of collaborator aliases to their Data Sharing Account Identifiers. Only users listed here can participate in the collaboration. Use the aliases defined here to refer to all collaborators, rather than using their data sharing account identifier directly. Must be unique in this collaboration and follow Snowflake identifier rules (maximum 25 characters).
analysis_runnersDescribes who can run an analysis in this collaboration. Each analysis runner is keyed by a unique alias. You must allow at least one account to run an analysis in this collaboration.
<analysis_runner_alias>Alias of account that can run an analysis in this collaboration. Alias is defined in the
collaborator_identifier_aliaseslist.data_providersData providers whose data this analysis runner can access. Each provider is keyed by the alias that is defined in
collaborator_identifier_aliases.data_offeringsA list of data offerings from this data provider that the analysis runner can access, or an empty array
[]as a placeholder so that data offerings can be added later. Each data offering is referenced by its ID, generated when the data provider calls REGISTER_DATA_OFFERING.
templates(Optional)The templates that can be used by this analysis runner. Each template is referenced by its ID. You can omit this in the initial spec, and still share templates with this analysis runner after the collaboration is created.
activation_destinations(Optional)Defines activation settings for the analysis results.
snowflake_collaborators(Optional)List of collaborators who can receive activated analysis results. Use the alias from the
collaborator_identifier_aliaseslist in this spec. All collaborators listed here must have the permissions described in Implementing activation.