Molted Email

About Molted Email

Molted Email is agent email with policy enforcement in the infrastructure rather than in the prompt, which is a direct answer to the failure mode its own testimonials describe: nothing ordinarily stops an autonomous agent from spamming your users. Every send is checked before it leaves — duplicate prevention on a dedupe key, cooldown windows, suppression lists, consent validation and rate limits — and inbound replies are scanned, including for prompt-injection attacks aimed at the agent reading them. Threads are tracked and classified by intent so the agent knows what it is answering. Around that sit journey orchestration, outcome attribution, deliverability monitoring and an operator console, all included on every tier including free; only volume changes between plans. Delivery routes across Amazon SES, Postmark and Resend with automatic failover to whichever provider is healthiest. Setup is a global CLI install and a signup command that returns a mailbox and API key, with SSE event streaming and magic-link generation for user sign-in flows. A Humanizer on the higher tiers rewrites AI-drafted mail in a chosen style and stores both versions so the change is auditable. Exceeding a limit returns a rate-limit error rather than an overage charge. Magic-link generation for user sign-in flows is a single command, and the operator console gives a human view over what the agents have been sending, which is the piece that makes the guardrails auditable rather than merely present.

Key Features

Tags