CLI Commands

The betat CLI is a thin dispatcher over Django management commands — every command below also runs as python manage.py <command> from framework/.

betat init

Declares this install’s community identity and writes its CommunityConfig — the one thing every other endpoint below depends on existing. Also runs an environment preflight (Python 3.11+, SQLite available), checks the declared community id resolves in DNS, collects an operator good-faith declaration + contact email (written to .env as an accountability record — never used operationally), and writes a standard manage.py to the working directory (a pip install gives you the importable betat_community package only, per DISTRIBUTION.md — betat init is the one guaranteed point with a real project directory to put it in). From there it’s plain Django: python manage.py migrate, createsuperuser, etc. work exactly as documented upstream. The declaration and email prompts are not skippable, even when the identity fields below are supplied as flags — this is deliberate: a script can’t fabricate a resolving domain or a human’s acceptance of the declaration.

$ betat init --id marinebiology-lagos.org --name "Marine Biology Lagos" \
    --domain "marine biology" --content-type scientific_observation \
    --store-uri https://marinebiology-lagos.org/betat/records \
    --auth-method community_peer_vouching
...
Do you accept this declaration? [yes/no]: yes
Contact email: operator@marinebiology-lagos.org
...
CommunityConfig written for 'marinebiology-lagos.org'.

Readiness:
  [x] identity declared — id=marinebiology-lagos.org
  [x] HI standard — human-originated, community-verified
  [x] auth method(s) — community_peer_vouching

Every flag can be omitted for a fully interactive run instead. --auth-method is repeatable to enable more than one authentication method — at least one from the protocol list is required (community_peer_vouching, cryptographic_signature, institutional_endorsement).

Database configuration

betat init declares community identity — the database engine is separate, set via the BETAT_DB environment variable and read directly by settings.py. No BETAT_DB set at all means SQLite (betat.sqlite3 in the project directory) — zero configuration, fine for evaluation. Point BETAT_DB at a PostgreSQL connection URL before a real deployment — no code changes, same installed package — see the Production Guide for the role setup PostgreSQL needs to actually enforce append-only at the database-permission level. Optionally set BETAT_DB_SCHEMA to isolate Betat’s tables in their own PostgreSQL schema instead of public — useful when sharing a database server with other applications; see .env.example for the exact steps.

betat runserver

Plain alias for manage.py runserver.

betat start

Alias for manage.py runserver, spelled as a verb a new operator reaches for first. Binds 0.0.0.0:8000 by default.

$ betat start
Starting Betat community server...
  Equivalent command: python manage.py runserver
  For production use: gunicorn betat_community.wsgi:application

betat backup

Backs up the configured database — a plain SQLite file copy, or pg_dump on PostgreSQL. No Django equivalent; run it from cron or a systemd timer for scheduled backups.

$ betat backup
SQLite backup written to: betat_backup_20260901_120000.sqlite3

betat check, manage.py check --deploy, manage.py shell

betat check is a plain alias for manage.py check. There is deliberately no betat-wrapped check --deploy or betat shell: Django resolves management commands by name, so a custom command sharing a built-in’s name (check, shell) replaces that built-in everywhere, and a wrapper that calls back into the same name via call_command() recurses into itself infinitely (see BLUEPRINT §1 Decision Log, 2026-08-30). Use the standard Django forms directly:

$ python manage.py check --deploy   # production-readiness checks (SECRET_KEY, DEBUG, HTTPS)
$ python manage.py shell            # interactive Python shell with the framework loaded

betat announce

Pings the registry: “new records available — crawl me now” (see API Endpoints for what communities publish; the registry itself is betat main’s job, not this package’s). Requires BETAT_REGISTRY_URL to be set — there is no default, since the registry’s own announce contract isn’t published yet.

$ export BETAT_REGISTRY_URL=https://registry.example.org/announce
$ betat announce
Announced 'marinebiology-lagos.org' to the registry.
Payload: {'community_id': 'marinebiology-lagos.org', 'store_uri': 'https://marinebiology-lagos.org/betat/records', 'announced_at': '2026-09-01T12:00:00Z'}

Can also fire automatically after every accepted submission — set BETAT_AUTO_ANNOUNCE=true (off by default; a slow or unreachable registry will never block or fail an accept, it just logs a warning).

betat export

Produces an integrity-verifiable dump of every provenance record this community has issued — every record’s record_id recomputes, and the whole bundle carries a export_hash so a reader can tell if the file itself was altered in transit (see core/export.py for exactly what “signed” means here — a recomputable hash, not a private-key signature; there is no community signing key in the seed implementation).

$ betat export --output betat-export.json
Exported 42 record(s) to betat-export.json

Omit --output to print the JSON to stdout instead — useful for piping straight into another tool.

{
  "betat_export_version": "0.1",
  "community": {"id": "marinebiology-lagos.org", "name": "Marine Biology Lagos", "...": "..."},
  "exported_at": "2026-09-01T12:00:00Z",
  "record_count": 42,
  "records": [ /* every record, exactly as PROVENANCE_SPEC.md defines it */ ],
  "export_hash": "sha256:9f4a2b..."
}