find out what you're in for._

$ curl -O https://pgrun.dev/discovery.sql
  discovery.sql              100%[==================>]  12.0K
$ shasum -a 256 discovery.sql
  bf33edf04e14da086437788eb7a851964103fc06312808380528f863976b1646  discovery.sql
✓ matches the hash printed below
$ psql -q -f discovery.sql "$DATABASE_URL"
$ ls -la pgrun-discovery.json
  -rw-r--r--  1 you  staff  9.2K  pgrun-discovery.json
✓ written locally. nothing was uploaded.
  drop it on this page to see your report.
$ 

discovery.sql captures the shape of your database — sizes, indexes, extensions, replication settings. Read-only SELECTs over the catalogs; it never reads a row of your data. Download it, read it, then run it:

$ curl -O https://pgrun.dev/discovery.sql
$ shasum -a 256 discovery.sql
# should print exactly:
# bf33edf04e14da086437788eb7a851964103fc06312808380528f863976b1646
$ psql -q -f discovery.sql "$DATABASE_URL"
#=> writes pgrun-discovery.json — open it, it's yours to read

Takes seconds on most databases, a minute or two on very large ones (it caps its own work and sets a 120s statement timeout). The hash catches a corrupted download. Works on anything psql can reach: RDS, Aurora, Heroku, Supabase, Neon, Cloud SQL, your own box.

Using a coding agent? Drop pgrun.dev/migrate.md into your repo and say "migrate my database to pgrun following this file" — it's the complete runbook: discovery, path choice, commands, cutover checklist, and hard safety rules (never writes to your source, never cuts over without your say-so).

Every step above is standard tooling — you can run it yourself, and the copy buttons hand each fix to your coding agent. Or we run it with you: standard migrations are included.