Proxy & data
Data lake API
Run read-only DuckDB SQL against an instance data lake through Blnk Cloud.
POST
Query ledger history (Parquet in the instance data lake) through Cloud. This is not the hot Data API, which reads the live instance database.
Use this for reporting and audits over a date range. For live lists, filters, and current wallet holdings, use the Data API.
For SQL tables, casts, and templates, see Insights.
Requires the
data:read scope. Production managed instances include a built-in data lake. See Managed instances.Request structure
string
required
Bearer token for an API key or OAuth access token with
data:read.string
required
Core instance whose lake you want to query.
string
Archive range start. Required if
end_date is empty. ISO date or datetime.string
Archive range end. Required if
start_date is empty.string
One read-only DuckDB
SELECT or WITH statement. Tables are inferred from the SQL. Do not send resource, resources, or org_id.string
Run a query saved in Insights instead of sending
sql.SQL response
array
Result rows. Column names match the SQL aliases.
integer
Number of rows returned in this run.
List without SQL
If you omitsql and saved_query_id, send a resource list instead. resource must be one of transactions, balances, ledgers, identity. Include start_date or end_date.
Saved queries
Saved lake queries live under/data/lake/queries.
Create body includes
instance_id, name, description, sql, range_mode (fixed or rolling), plus either start_date / end_date or rolling_days.