AI assistants and search decision
Saner.AI
A capable developer can implement a useful ADHD-focused assistant (chat + tasks + calendar + notifications) within a week and maintain it; the hosted product's polish, user base, and any proprietary models would be lost.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- A minimal replacement provides a conversational assistant that stores and surfaces short-term tasks/reminders, syncs or imports calendar events, accepts plain-text or web input, and sends scheduled notifications.
What it still won’t have
- Polished UX and mobile app polish
- Any proprietary models, training data, or closed-source optimizations
- Existing user base, analytics, and product support
- Platform-level reliability and SLAs
What remains hard
- Product polish and ongoing maintenance
First-year cost
No published price
Saner.AI does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal AI personal assistant web app using Next.js (React) frontend, FastAPI backend, and Postgres. Core features: chat interface that sends messages to an LLM (configurable OpenAI-compatible API), CRUD task/reminder storage, one-way Google Calendar import (OAuth and event import), background scheduler to deliver notifications via an email provider (SendGrid) and SMS (Twilio), and basic auth (magic-link). Out of scope: native mobile apps, advanced agent orchestration, training custom models, and multi-tenant admin UI. Include error handling for network, API quota, and scheduler failures, automated unit tests for backend endpoints, and deployment scripts (Dockerfiles and a simple Terraform or Docker Compose setup).
How we checked
How the score was reached
- Build verdict base78
- An open-source build was found+5
- 3 cited sources+3
- Evidence score86
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 3
Every page the run actually retrieved.
- official productSaner.AI - AI Personal Assistant for ADHD | Your Jarvis is here
- open sourceleon-ai/leon
- open sourceagentscope-ai/QwenPaw
Integrity checks
What held up, and what did not.
