Smart Email Client

Part of the worlds-biggest-software-project initiative.

An open-source, AI-native email client that delivers prioritisation, drafting, and follow-up automation across any provider — without locking users into Gmail, Outlook, or a single LLM vendor.

Smart Email Client is a candidate project for a production-grade, provider-agnostic email client built around modern AI capabilities. It targets knowledge workers, customer-facing teams, and privacy-conscious professionals who are underserved by today's Gmail-only or expensive proprietary tools.

Why Smart Email Client?

Leading AI clients like Superhuman ($30/month) and Serif ($29/month) are expensive and Gmail-only, leaving Outlook and IMAP users as second-class citizens.
Shortwave, Gmelius, HiverHQ, Notion Mail, and Serif are all tied to Gmail; cross-provider AI parity does not exist in the market.
Privacy-first options like Canary Mail run AI on-device but lag cloud-based competitors in capability — the gap between privacy and power has not been closed.
AI features are overwhelmingly English-optimised; non-English professionals receive shallow drafting and summarisation quality.
Mail-0 / Zero is the only open-source AI email client today, but is early-stage with no production-grade autonomy features. A full-featured open-source alternative to Superhuman does not yet exist.

Key Features

Universal Inbox and Triage

Single OAuth connection across Gmail, Microsoft 365, and any IMAP provider
AI priority scoring with user-trainable approve/dismiss signal
Smart sorting and auto-labelling derived from intent and content
Snooze and follow-up reminders with AI-detected due dates extracted from thread content

AI Drafting and Summarisation

One-sentence thread summaries surfaced above each conversation before it is opened
Reply drafting from short user prompts or full thread context
Voice and style learning so drafts match the user's personal writing patterns over time
Autonomous background drafting mode that pre-drafts replies before the user opens a thread
Attachment summarisation reading PDFs and Word documents and surfacing key points inline

Cross-App Context

Pull relevant Notion pages, Slack threads, and calendar events into the compose panel
CRM-aware drafting that synthesises deal history, last touch, and open tasks without switching apps
Email-to-task conversion and push to external project management tools
Natural-language workflow rules connecting email to third-party services

Privacy and Compliance

Optional on-device AI processing for privacy-sensitive users
PII anonymisation layer before any LLM API call
TLS in transit by default; optional OpenPGP end-to-end encryption
Multilingual AI support targeting English, Spanish, French, German, and Japanese at minimum

Team and Automation (v1.1+)

Shared inbox with assignment, internal notes, and SLA timers
Kanban / board view converting an inbox into a project pipeline
AI-generated meeting scheduling that learns user preferences and confirms slots
MCP server exposing the inbox as a tool context for external AI agents

AI-Native Advantage

Unlike incumbents that bolt LLM features onto a fixed UX, Smart Email Client treats AI as a first-class layer: contextual reply generation reads the full thread, attachments, and CRM record before drafting; adaptive priority scoring learns individual urgency signals rather than relying on generic heuristics; and proactive follow-up orchestration detects sent messages awaiting a reply and surfaces or auto-sends a chase note. Tone and compliance guardrails for regulated industries flag or rewrite drafts before send, closing a gap that no current client covers end-to-end.

Tech Stack & Deployment

The project targets self-hosted, cloud, and hybrid deployments. Universal provider support is built on IMAP, SMTP, MIME (RFC 2045–2049), OAuth 2.0, and OpenID Connect, with DMARC / DKIM / SPF preserved for sender alignment and CalDAV / CardDAV used for calendar and contact integration. The Mail-0 / Zero MIT-licensed reference implementation (Next.js, React, TypeScript, TailwindCSS, Shadcn UI, PostgreSQL, Redis) demonstrates a viable open-source baseline. A pluggable LLM backend allows users to choose between hosted models (GPT-4, Claude, Gemini) and local models via Ollama or llama.cpp.

Market Context

The AI email management category serves tens of millions of professionals globally and is growing in double digits as AI features become table stakes. Consumer pricing runs $7–$17/month, professional tiers cluster at $25–$30/month, and enterprise seats trend $12–$20/seat/month. Primary buyers are individual knowledge workers drowning in inbox volume, sales and customer-success teams needing fast response times, executives wanting high-signal inbox views, and distributed teams sharing inboxes. Candidate signals: complexity 6/10, demand High, domain availability High.

Project Status

This project is in the research and specification phase.
Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.