Skip to main content
An empty test database doesn’t catch real bugs. A database full of copied production data is a compliance problem and a security review you have to sit through. Flock is the third option: point it at your schema and it generates a synthetic population that fits your tables, keeps every foreign key valid, and contains no PHI because no real record ever entered the pipeline.

​
One loop: schema in, seeded database out

Flock reads your database schema, a live connection or a SQL DDL file, and maps the foreign keys into a dependency graph. It generates records in that dependency order so every reference resolves. You get the population as FK-ordered SQL INSERTs, CSV, HL7 v2 message streams, or FHIR transaction Bundles, and you can seed a connected database directly.

​
Why the data looks real

The generated cohort isn’t uniform noise. Demographics track US Census age, sex, race, and geographic distributions, and clinical content is grounded in a 108-condition comorbidity matrix built from public epidemiology (CDC WONDER, NHANES, CDC surveillance). Diabetes pulls in hypertension, obesity pulls in sleep apnea, and temporal order holds across the record: admissions precede discharges, lab orders precede results, medication start dates precede end dates. That grounding is why a QA run against Flock data surfaces the failures a hand-built handful of test patients never will. It is a credibility proof-point, not the headline. The headline is schema-aware, FK-safe, no-PHI data in your target format.

​
When something fails

​
Getting Flock

Flock is a paid product and requires a Pidgeon account sign-in. Post’s free CLI (message generation and validation) does not include Flock; the flock command is delivered through the Flock desktop app and the Pro CLI. See Get started with Flock.

​
Next steps

Schema Intelligence

How Flock classifies tables and maps foreign keys.

Population Generation

Demographics, correlated conditions, and temporal coherence.

Output Formats

SQL, CSV, HL7 v2, and FHIR Bundles.

Seed a database

From an empty schema to a seeded population.