The agent-mail vocabulary.

The behavioral IP segregation model that routes sends between behavioral pools — Clean and Mixed today, an explicit Outbound pool in Phase 2 — based on the classifier's read of the send's behavioral signals, not account settings.

The behavioral model that decides which pool (Clean / Mixed / Outbound) a send goes to. Reads volume curve, recipient sourcing, engagement signals, content patterns, and the per-agent baseline.

0–1 risk score on every inbound structured reply event. Output of Mails.ai's six-category prompt-injection scanner. Above 0.95 the event is flagged `quarantined`.

Mails.ai's drop-in Model Context Protocol server. Distributed via npm as @mailsai/mcp-server. Exposes five tools (send, on_reply, list_threads, suppress, get_reputation) to MCP-capable runtimes — Claude Code, Cursor, Cline, Continue, Windsurf.

Mails.ai's usage-based pricing tier, coming soon — Stripe metered billing at $0.001/send + $0.002/inbound (+$0.003 opt-in classify). No monthly minimum. Aligns cost to actual agent send + inbound volume rather than over-committed monthly tiers.

An agent's reputation score (0–1), built from real engagement. Queryable today via the mails.get_reputation tool; the current API returns an agent's own score. Designed to propagate across the network as the cohort grows, so a banned sender loses standing across every account that has joined.

A structured object representing an inbound email reply. injection_score + sender_reputation are computed on every event; intent, entities, and urgency are added when opt-in classification is enabled.