This is the vertical I was most skeptical about before I shipped it. Family calendar apps are a graveyard — Cozi, TimeTree, FamilyWall, half a dozen Kickstarters that didn't make it past year two. Most of them pivot toward ads, or subscriptions, or social features that weren't what anyone asked for. The core problem (everyone in the house is on a slightly different schedule) is genuinely hard to solve at scale, which is why nobody's solved it.

familyhub doesn't try to solve it at scale. It solves it for one family, in their own house, on their own laptop. That's the whole trick.

What's In It

Five core features and nothing else:

Why I Built This

Jess and I have one kid in elementary, one in middle school, and a running joke that our family's actual ops manager is the shared Google Calendar. The calendar works for adult things (work, appointments, travel). It fails for family things — chore rotations, allowance, the running grocery list, who's driving whom to Tuesday's soccer practice.

We tried Cozi. We tried TimeTree. Both had ads. Both wanted our data. Both kept pushing "upgrade to premium" at us. I don't want my family's schedule being ad inventory. Full stop.

familyhub runs on our home server and serves the family over the Wi-Fi. No cloud. No ads. No signup. Anyone in the house opens a bookmark and they're in. The kids have their own colored column. The Costco list is always current. The babysitter gets a printable page.

The Build

Flask + SQLite + Jinja. One 900-line Python file for the routes. HTMX for the interactive bits (check off a chore, add a grocery item). A Python script that generates iCal feeds per-person so each family member's phone calendar picks up their relevant events.

git clone https://github.com/Dangercorn-Enterprises/familyhub.git
cd familyhub
pip install -r requirements.txt
echo "AUTH_PASSWORD=household" >> .env
python app.py
# → http://localhost:8427 (bookmark this in the kitchen Kindle)

Deterministic port 8427. I've got mine running on a Raspberry Pi 4 with a static DHCP lease. Total hardware cost: $45. Total monthly cost: the power draw of a Raspberry Pi, which is measured in cents.

Auth for a Household

Single password. We share it with the whole family. For kid-specific views (their chore list, their allowance), we don't do per-user auth — we use scoped bookmarkable URLs with a token in the query string. Each kid has their own link saved on their tablet home screen.

This sounds insecure. It's not, for this context. There's no sensitive data leaving the house. The server isn't internet-facing. The trust boundary is "who's on the home Wi-Fi." That's your family, your babysitter, and whoever you let on your network. Appropriate for the threat model.

Pricing

$9/mo for Hosted is intentionally cheap. This is a family tool; it should be cheaper than one streaming service. The margin on the hosted tier is meant to subsidize ongoing development of the free self-host tier, not to be a profit center.

Who It's For

Two-parent households with 2-4 kids. Single-parent households with any number of kids. Multi-generational households (this is where the Extended Family tier starts to matter). Anyone who's tried two family-calendar apps and given up.

Not for: single people, empty-nesters, couples without kids. You don't need this. A regular calendar is fine for you. This is specifically for the coordination complexity of a household with children.

Your family's schedule should not be ad inventory. That's not a technical statement. That's a values statement. We built the app accordingly.

Related

childmilestone is the natural complement — familyhub handles the week; childmilestone handles the memory. bookcircle for family book-club use. cheesemaking if your family's weird the same way mine is.

Repo: github.com/Dangercorn-Enterprises/familyhub. Clone it, run it, use it for a month. Email me what would make it better.