MailMate AI Desktop — Open-Source Email Assistant That Summarizes, Prioritizes, and Creates Tasks Locally

Description and use

MailMate AI Desktop is a local‑first Gmail assistant that summarizes emails, classifies them by priority, and creates tasks, sending notifications to Telegram. All models run locally, ensuring privacy and zero cloud costs.

Problem

Most people read every email to decide what matters, creating a productivity tax.

Outcome

Processing 10 real inbox emails produced AI summaries and classified priorities; a P0 security breach task was auto‑created with a calendar event, a P1 design review task awaited confirmation, and other emails were summarized only.

Workflow

• Read Gmail (read‑only) → Normalize email → Summarize with Qwen2.5‑0.5B via vLLM → Classify summary with Laya typed‑decision engine → Apply application‑side policy → Persist approved actions in SQLite → Notify via Telegram bot → (optional) create Google Calendar event

Evidence

MailMate AI Desktop flips this: two open-weight models read your Gmail first, summarize every email into 1–2 sentences, classify it by priority, and deliver only what needs your attention — directly to your Telegram.

Source

The summarizer runs first. Laya then classifies the summary — not the raw email. Cleaner signal, better decisions.

Source

Key safety rule: A model prediction alone never authorises anything. An explicit policy module enforces the rules: ambiguous deadlines are never auto‑scheduled, P1/P2 tasks require your confirmation, calendar events are always a separate explicit step.

Source

Zero running cost. Both models run locally on a MacBook Intel — no API bills, no per-token charges.

Source

Your inbox never touches a third‑party server. OAuth tokens stay on disk. The SQLite database is local.

Source