An always-on second brain you talk to. Voice notes in Telegram → typed, linked knowledge in your Obsidian vault. Runs 24/7 on the Claude subscription you already have.
Python ai-agentsai-automationclaudeclaude-codeebbinghausknowledge-graphllmmemory-systemobsidianpersonal-assistantpkmproductivitypythonsecond-braintelegram-botvoice-assistant
"""Contract tests for the install/upgrade scripts.
setup.sh once generated the dead v2 layout (system-level d-brain-* units)
— a fresh install was broken. These pins keep the install path honest:
setup.sh = interactive questions only, upgrade.sh = the single source of
truth for services and health.
"""
from pathlib import Path
ROOT = Path(__file__).parent.parent
SETUP = (ROOT / "setup.sh").read_text()
UPGRADE = (ROOT / "upgrade.sh").read_text()
def test_setup_delegates_services_to_upgrade():
assert "upgrade.sh" in SETUP
def test_setup_has_no_dead_v2_layout():
assert "d-brain-" not in SETUP # legacy unit names
assert "/etc/systemd/system" not in SETUP # v3 uses systemd --user
def test_setup_login_check_uses_json_not_prose():
# `claude auth status | grep "Logged in"` broke when the CLI changed
# its prose; the JSON field is the stable contract.
assert "loggedIn" in SETUP
def test_setup_asks_timezone():
assert "TZ=" in SETUP
def test_upgrade_targets_v3():
assert "v3.1" not in UPGRADE
assert "dbrain-bot.service" in UPGRADE
def test_legacy_mac_installer_is_gone():
assert not (ROOT / "install.sh").exists()