Blindata Agent connects your infrastructure to the data catalog. Through connectors—specialized extraction jobs configured and scheduled from Blindata—you can harvest physical entities, fields, routines, dataflows, and glossary links from systems that would otherwise require manual documentation.

Most connectors run on a schedule: the agent reaches the source, extracts metadata, and uploads it to Blindata. Jobs can also be triggered on demand and monitored from the agent history.

Databases and generic sources

Relational databases and JDBC-compatible systems are the most common crawling targets. On the CRAWLING tab of a system detail page you choose between two strategies:

With the JDBC driver strategy, the agent reads metadata through the database’s JDBC driver. Default extraction queries transform the driver output into catalog assets. This is the straightforward choice when INFORMATION_SCHEMA or equivalent catalog views are available and performance is acceptable. Configuration details and default queries are in Import Metadata with SQL or JDBC .

With the SQL strategy, you write the extraction queries yourself. Each query maps rows from source tables or views to a specific Blindata resource type—physical entities, fields, constraints, routines, dataflows, semantic links, and more. This interface tables approach is the right fit when metadata lives in custom schemas, lineage repositories, staging tables, or any structure that standard catalog views do not cover. The pattern, supported query types, and when to use them are explained in Metadata import with interface tables . Column aliases, job setup, and per-database examples are in Import Metadata with SQL or JDBC .

For a walkthrough of the crawling UI in Blindata, see Automated Metadata Crawling .

Application connectors

When the source is a SaaS or analytics platform rather than a database you query directly, Blindata provides dedicated connectors that call the platform’s APIs and map the response to catalog assets—often including lineage where the platform exposes it.

Power BI imports workspace, dataset, and report metadata through the Power BI REST APIs, including .pbix enrichment for report-level detail and dataflow generation from model dependencies.

Azure Analysis Services imports tabular model metadata by analyzing .bim files uploaded from an AAS instance, including entities, measures, and lineage into the data catalog.

Salesforce imports standard and custom object metadata and their relationships through the Salesforce REST APIs, populating a Salesforce system in the catalog automatically.

Custom developments

When none of the built-in connectors fit, the agent can orchestrate external code you own. Custom Jobs let you register an HTTP endpoint, pass parameters and connections in JSON, and trigger the job on demand or on a schedule—useful for proprietary extractors, internal tools, or integrations that are not yet first-class connectors.

Definitions