The problem
Email volume past ~50 messages/day exceeds what most people can triage well. Hosted services like Superhuman or HEY work but ship your inbox to a third party. Self-hosted lets you keep the data while still gaining the leverage.
Recommended setup
| Agent | Hermes Agent |
| Hardware | Raspberry Pi 5 (8 GB) or Hetzner CX22 (€5/month) |
| LLM | Claude 4.5 Sonnet via Anthropic API (or OpenRouter) |
How it works
Hermes Agent connects to your IMAP/JMAP inbox via a custom MCP tool. The tool fetches headers + first 500 characters of new mail, the agent classifies (action / reply / archive / spam) and either replies (drafts saved to drafts folder, never auto-sent) or files. Approval flow on every send.
Reality check
After three months of running this on my own inbox: 80% classification accuracy on routine mail, ~30 false positives per month I had to correct. Time saved: 4-5 hours per week. Anthropic API cost: ~€18 per month for ~3,000 messages classified. Cheaper than Superhuman, less polished UI, but the data stays on my hardware.
What breaks
- Long, threaded conversations (agent loses context past ~10 turns)
- Highly contextual replies that need knowledge from outside the email itself
- Anything legal, financial or relationship-sensitive (always reply yourself)
Alternative setups
Nanobot + custom IMAP tool if you want to read every line of code that touches your inbox. ZeroClaw + local Mistral 7B if you can tolerate ~62% classification accuracy in exchange for cloud-free operation.