# RadMail > Your inbox, triaged by what actually matters. RadMail is an email-importance engine for business inboxes: two-axis triage (importance x urgency), an answer-ready inbox (important mail surfaces with a reviewable draft prepared), commitment follow-through in both directions, and a verifiable BEC hard-stop (money / changed-banking / first-contact stays human-only, forever). ## Public v1 REST API Base URL: https://app.radmail.ai/api/v1 Auth: Authorization: Bearer tmk_live_... — mint a key at https://app.radmail.ai/settings/api-keys Scopes: read (all GETs) · write (POST /ingest, POST /send, POST /send/{id}/confirm) Tiers: Pro and up: full access at 120 requests/min per key. Free and trial workspaces: every read endpoint works on the metered free tier at 100 requests/day per key; write requires Pro+ (403 api_not_entitled_upgrade_to_pro). Rate headers: Every response carries X-RateLimit-Limit / X-RateLimit-Remaining / X-RateLimit-Reset; 429s add Retry-After. Errors: Errors are { ok: false, error } — 401 missing_or_malformed_api_key | invalid_api_key, 403 missing_scope: | api_not_entitled_upgrade_to_pro, 429 rate_limited. Paginated endpoints take limit + offset and return { data, pagination: { limit, offset, total, hasMore } } (default limit 25, max 100 unless noted). ### GET /right-now (read, free tier OK) What can't be missed right now — the inbox's own hero lane, ranked by the two-axis importance engine. Each item carries importance (0–100), urgency (0–100), band (low | normal | high | critical) and plain-English reasons. Candidate pool is recent mail (last 90 days, unarchived, not spam); `total` is qualifying rows in that pool. Params: limit (query) — default 10, max 50; offset (query) — default 0 Returns: { ok, data: [{ id, receivedAt, from, fromName, subject, classification, importance, urgency, band, reasons[], counterparty, threadId }], pagination } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/right-now?limit=10" ### GET /search (read, free tier OK) Ranked search over your emails — every term must match sender, subject, or body; header matches rank above body-only matches. Hits include a ~160-char body snippet and matchedIn (which fields hit). Params: q (query, required) — search terms (max 200 chars, first 8 whitespace-separated terms used); from (query) — substring filter on sender email or name; after (query) — ISO date — only mail received after; before (query) — ISO date — only mail received before; limit (query) — default 25, max 100; offset (query) — default 0 Returns: { ok, data: [{ ...email metadata, snippet, matchedIn[] }], pagination } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/search?q=invoice+overdue" ### GET /emails (read, free tier OK) List your emails newest-first (metadata only — bodies stay out of list views). Params: limit (query) — default 25, max 100; offset (query) — default 0 Returns: { ok, data: [{ id, receivedAt, from, fromName, subject, classification, classificationSource, isSpam, needsOwnerEyes, counterparty, threadId }], pagination } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/emails?limit=25" ### GET /emails/{id} (read, free tier OK) Fetch one email by id, including the text body and attachment count. Params: id (path, required) — email UUID from a list/search response Returns: { ok, email: { ...metadata, to, textBody, attachmentCount, llmClass } } — 404 if not found Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/emails/" ### GET /emails/by-message-id (read, free tier OK) Resolve an RFC 5322 Message-ID to your email's backend id. Lets a client map ANY local thread — including mail the engine didn't surface — to its RadMail id, e.g. to submit a 👍/👎 via POST /feedback. Returns the most recent match when a Message-ID repeats. Params: messageId (query, required) — the RFC 5322 Message-ID header value (max 998 chars, angle brackets included as sent) Returns: { ok, data: { id, messageId, from, fromName, subject, receivedAt, classification, isSpam, threadId } } — 404 not_found Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/emails/by-message-id?messageId=%3Cabc%40mail.example.com%3E" ### GET /prepared-drafts (read, free tier OK) The answer-ready inbox, as data: your stored AI-prepared reply drafts, newest first, each linked to its source email. Read-only — a draft is NEVER auto-sent; a human approves every reply (the BEC hard-stop stays sacred). Filter to one email with ?emailId=. Params: emailId (query) — email UUID — return only drafts prepared for that email; limit (query) — default 25, max 100; offset (query) — default 0 Returns: { ok, data: [{ id, emailId, style, model, promptVersion, createdAt, body, from, subject }], pagination } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/prepared-drafts?limit=25" ### GET /commitments (read, free tier OK) Open commitments in both directions — what you owe and what you're owed — extracted from mail and tracked to completion (north star: nothing lost, in either direction). Params: limit (query) — default 25, max 100; offset (query) — default 0 Returns: { ok, data: [{ id, direction, party, action, actionType, dueDate, duePhrase, state, confidence, counterpartyEmail }], pagination } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/commitments" ### GET /usage (read, free tier OK) Work-units rollup: how much work RadMail did for your org over a window — prepared reply drafts, lead dossiers, and supporting triage calls, grouped by feature, with honest engine cost. Params: days (query) — window, default 30, max 90 Returns: { ok, windowDays, workUnits, triageCalls, totalCalls, totalCostCents, byFeature[] } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/usage?days=30" ### POST /ingest (write, Pro+) Push an email into your workspace from your own infrastructure. Runs the full pipeline — classification, threading, importance, commitments, webhooks — exactly like every other inbound path. The org is bound to your key; the body can never assert a different tenant. Params: fromEmail (body, required) — sender address; subject (body) — subject line; rawText (body) — plain-text body; rawHtml (body) — HTML body; toEmail (body) — recipient address; fromName (body) — sender display name; receivedAt (body) — ISO timestamp, defaults to now; idempotencyKey (body) — dedupe key — safe retries return 200 instead of creating a duplicate; headers (body) — object of raw headers; attachments (body) — array of attachment descriptors Returns: 201 on create, 200 when deduped by idempotencyKey; { ok, ... } Example: curl -X POST -H "Authorization: Bearer $RADMAIL_API_KEY" -H "Content-Type: application/json" -d '{"fromEmail":"vendor@example.com","subject":"Invoice #1042","rawText":"Net-30, due Friday."}' "https://app.radmail.ai/api/v1/ingest" ### POST /feedback (read, free tier OK) Teach the importance engine: a 👍/👎 on one of your emails records the operator's ground-truth label (important | not) that the tuner + scorer consume, so the engine learns what actually matters to you. Read-scope — it's a self-org training signal on your own RLS-scoped mail, not a privileged write. One current vote per email; a repeated vote corrects rather than piles up. Params: emailId (body, required) — email UUID (from any list/search response, or GET /emails/by-message-id); verdict (body, required) — 'up' (👍 important) or 'down' (👎 not) — anything else is 400 invalid_verdict Returns: { ok, label: 'important' | 'not' } — 404 email_not_found if the id isn't yours Example: curl -X POST -H "Authorization: Bearer $RADMAIL_API_KEY" -H "Content-Type: application/json" -d '{"emailId":"","verdict":"up"}' "https://app.radmail.ai/api/v1/feedback" ### POST /send (write, Pro+) Human-initiated send with the outbound double-check gate: RadMail runs deterministic checks (recipient-typo distance, placeholder scan, attachment presence, reply-context match) plus an LLM review BEFORE anything leaves via your connected M365 mailbox. Clean sends to normal recipients release immediately (201). Anything questionable — and ALWAYS regulator/government/court/bank recipients or money/banking-change/first-contact content — returns 202 held with a one-time confirmToken: a human re-reads the rendered email and confirms. Feature-flagged (404 send_api_disabled until enabled); the from address must be one of your org's connected M365 mailboxes with send permission. Params: from (body, required) — a CONNECTED M365 mailbox address of your org (the send goes out through it); to (body, required) — recipient address or array of addresses; cc (body) — cc address(es); bcc (body) — bcc address(es); subject (body, required) — subject line (an empty subject holds the send); markdown (body, required) — markdown or plain-text body (aliases: body, text) — rendered to safe HTML server-side; inReplyTo (body) — Internet Message-ID being replied to — enables the reply-context match check + threading; attachments (body) — array of { filename, contentType, contentBase64 } — max 10, 3 MB total decoded; requestedBy (body) — who is asking (e.g. 'doug-claude-session') — recorded as send provenance Returns: 201 { ok, status: 'sent', requestId, sendId } · 202 { ok, status: 'held', requestId, tier, holdReasons[], findings[], renderedHtml, confirmToken, confirmExpiresAt, confirmEndpoint } · 409 mailbox_send_scope_missing when the mailbox was connected read-only (reconnect with send permission) Example: curl -X POST -H "Authorization: Bearer $RADMAIL_API_KEY" -H "Content-Type: application/json" -d '{"from":"you@yourco.com","to":"kat@vendor.com","subject":"Thursday delivery","markdown":"Confirming Thursday works."}' "https://app.radmail.ai/api/v1/send" ### GET /send/{id} (read, free tier OK) Read one send request — the human re-read surface before confirming a held send: status, tier, hold reasons, every deterministic finding, the LLM review verdict, and the rendered final HTML exactly as it would leave. Params: id (path, required) — send request UUID (from POST /send) Returns: { ok, request: { id, status, tier, from, to, cc, bcc, subject, renderedHtml, findings[], holdReasons[], llmVerdict, sendId, ... } } Example: curl -H "Authorization: Bearer $RADMAIL_API_KEY" "https://app.radmail.ai/api/v1/send/" ### POST /send/{id}/confirm (write, Pro+) Release a HELD send after a human re-reads it — the second signature on the double-check gate. Requires the one-time confirmToken from the hold response; the release claim is atomic, so a racing double-confirm can never double-send. Tokens expire after 24 hours (410 confirm_token_expired — re-submit the send). Params: id (path, required) — send request UUID; confirmToken (body, required) — the snd_cfm_… token returned when the send was held (shown once, never stored); confirmedBy (body) — who confirmed (recorded on the request) Returns: { ok, status: 'sent', requestId, sendId } — 403 invalid_confirm_token, 409 already_sent, 410 confirm_token_expired Example: curl -X POST -H "Authorization: Bearer $RADMAIL_API_KEY" -H "Content-Type: application/json" -d '{"confirmToken":"snd_cfm_...","confirmedBy":"doug"}' "https://app.radmail.ai/api/v1/send//confirm" ## MCP (for AI agents) The RadMail MCP server (npm: radmail-mcp, MIT) speaks this API for you — zero-auth sandbox out of the box; set RADMAIL_API_KEY for read-only access to your real inbox. Run: npx radmail-mcp ## Docs - https://app.radmail.ai/docs/api — full API reference (this content, human-readable) - https://app.radmail.ai/demo — live demo inbox running the real scoring engine, no login - https://app.radmail.ai/trust — security posture, subprocessor register, verifiable refusal receipts - https://radmail.ai — product site