Head to head · Agent inboxes · October 2026 research run

AgentMail API + MCP vs Lumbox

AgentMail API + MCP scores 72.8 (BB) on agent readiness against Lumbox's 60.7 (C), and leads in 5 of 7 scored categories. Lumbox leads on security & auth and transparency & trust. Both do agent inboxes. AgentMail API + MCP accepts x402. Lumbox doesn't.

Best email inbox APIs for AI agents · All 66 inboxes comparisons

Which one, for what

AgentMail API + MCP BB

Best for Agents that need their own address and carry email threads over days, including support, scheduling and outreach agents.

Ahead on

  • Reliability, 61 against 45
  • Schema & documentation, 88 against 82
  • Agent ergonomics, 86 against 78
  • Payments & pricing, 85 against 40
  • Maintenance & community, 87 against 23

Also in its favour

  • Agent-ready, a grade of BB or better
  • An agent can pay per call over x402, with no account

Watch for

Email sending was down for 8 hours 7 minutes on 19 August 2026

Lumbox C

Best for Agents that need their own address to receive sign-up codes and links and to send replies.

Ahead on

  • Security & auth, 71 against 65
  • Transparency & trust, 72 against 65

Also in its favour

  • No incidents deducted, where AgentMail API + MCP loses 3 points for them

Watch for

No status page, incident history or SLA; the terms say availability isn't guaranteed

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Maintenance & community AgentMail API + MCP +64

AgentMail API + MCP 87

Changelog entry on 28 October (inbox signatures) and MCP server commits on 29 October (30). 19 dated changelog entries since 20 July, eight of them after 6 October (20). Closed service with a public changelog, Discord on the free plan and email support from Developer up. We didn't test support (12 of 15). to.agentmail/agentmail is active in the official registry under a DNS-verified namespace, and the TypeScript and Python SDKs are current (15). The MCP server pins agentmail 0.5.49 and agentmail-toolkit 0.18.0 since 29 October, patched its runtime dependencies on 26 October and runs contract tests and security workflows in CI. npm couldn't be read on 1 November to confirm those are the latest (10 of 10). 30+20+12+15+10 = 87.

Lumbox 23

The latest package releases found are lumbox 0.3.1 on npm (19 June 2026) and @lumbox/mcp-server 0.2.2 (12 June 2026), 140 days before this check, the 91 to 180 day band (10). No releases or dated changelog entries in the 90 days from 8 August (0). A closed service with a support address and an X account; no changelog (5 of 15). Not in the official MCP registry (search for lumbox returned nothing on 6 November); Node and Python SDKs exist but haven't changed since June (5 of 15). The MCP server card says 0.2.1 while npm has 0.2.2, and the SDK repository named in npm isn't publicly readable (3 of 10). Total 23.

Payments & pricing AgentMail API + MCP +45

AgentMail API + MCP 85

x402 on AgentMail's own x402.api.agentmail.to and x402.ws.agentmail.to, USDC on Base, Polygon or Solana. Only inbox creation has a published price ($2), and the 402 challenge on 1 November still names api.paysponge.com and asks 2,000,000 USDC base units on Base, Solana, Avalanche, X Layer and Polygon, so a third-party layer sits in front (30 of 40). Plan prices are public with per-unit add-ons, $2 a month per extra inbox, domain or 1,000 emails and $5 a month per extra seat (15 of 20). Free plan with 3 inboxes and 3,000 emails a month, 'No credit card required' (20). An agent can pay by x402 with no account, or sign itself up by API and get a key, with full access after a person confirms an OTP (20).

Lumbox 40

No x402, MPP or L402 (0). Four plans with prices and limits published without login, no per-email price (10 of 20). Free plan with 3 inboxes, 500 received and 100 sent emails a month, no card (20). Agent self-signup lets an agent create its own key, which a person approves once in a browser (10 of 20). Total 40.

Reliability AgentMail API + MCP +16

AgentMail API + MCP 61

Better Stack status page with components for the us-east-1 and us-west-1 APIs, email sending and the console, and incident history (20). One major outage in the 90 days to 1 November, email sending down for 8 hours 7 minutes on 19 August 2026, which stays in the window until 17 November, plus a 5-minute sending blip on 9 September and nothing since. The MCP repository's own incident write-up describes the hosted MCP server timing out for most of 19 and 20 August, and the status page still has no MCP component (10 of 30). Daily sending caps are published per plan (Free 100 a day, Developer 1,000 a day, Startup 15,000), but API request limits are only called 'generous' (8 of 15). 429s carry Retry-After, usually one second, plus message and fix fields, the SDKs retry on their own, and clientId covers idempotent creates (15). The pricing table marks an SLA as Enterprise only, with no terms published (0). The core API has no beta label, and Calendar left AgentMail Labs on 19 October, but the API lives under /v0 and Agent Armor is still in Labs (8 of 10).

Lumbox 45

Graded with the hosted lines. No status page was found; none is linked from the site or llms.txt, and status.lumbox.co doesn't resolve (0). With no history to read, the incident line takes 5. 120 requests a minute per API key, per IP for unauthenticated routes, with X-RateLimit headers (15). 429 returns retry_after_seconds, the spec gives a backoff schedule and says which statuses not to retry, and Idempotency-Key works on every POST for 24 hours (15). The terms say Lumbox doesn't guarantee uninterrupted service, and no SLA was found (0). The core API carries no beta label; dedicated phone numbers are early access (10). Total 45.

Agent ergonomics AgentMail API + MCP +8

AgentMail API + MCP 86

Over REST, lists take limit and pageToken, and extracted_text strips quoted history so an agent reads only the new reply. The hosted MCP server now lists 51 tools on API-key sessions and 53 on OAuth sessions, after the calendar tools were added on 19 October. Names, descriptions and input schemas come to about 77,600 characters (roughly 19,400 tokens), output schemas add about 138,000 more, and only the stdio bridges can filter with --tools. The award stays as read on 1 and 6 October (15 of 25). Thread and message lists filter by labels, senders, recipients, subject, dates, spam and trash, and since toolkit 0.17.0 (28 October) blocked and unauthenticated mail too (20). Errors name the limit and the fix (18 of 20). Inbox creation takes a clientId for idempotent retries, a repeated upload_file call returns the same file, and every MCP tool carries readOnly, destructive, idempotent and openWorld hints. Sends have no idempotency key (18 of 20). Official SDKs for TypeScript and Python, and only inboxId and to are required to send (15).

Lumbox 78

18 MCP tools, the 11 to 30 band; OAuth scopes narrow what a grant can do, but no read-only tool subset was found (15 of 25). Cursor pagination with limit and has_more, search, and an SSE stream for new mail (20). Errors are a human-readable message with the HTTP status, plus retry_after_seconds on 429 and written guidance on what to retry (12 of 20). Idempotency-Key on every POST; MCP annotations weren't readable without a session (16 of 20). One POST creates an inbox with a default address; official Node and Python SDKs (15). Total 78.

Transparency & trust Lumbox +7

AgentMail API + MCP 65

The API is closed under published terms. The hosted MCP implementation is open source under MIT (18 of 30). The privacy notice was replaced on 12 October 2026 and is unchanged on 1 November. It now keeps personal information 'for as long as you use the Services, or as long as necessary' with no periods, where the 27 September version gave mail until deleted, backups 35 days, metrics 90 days and logs 365 days. It now lists 'training, and fine-tuning our models' using Email Content 'except where we have agreed otherwise in a written agreement', where the earlier version said email content isn't used to train AI models. The security FAQ agrees with the new notice and says a DPA and the optional 24-hour message expiry are available on request. No published DPA, as openhands was read for retention 'as long as necessary' and training on content (12 of 30). No deprecation policy or dated notices found (5 of 20). The subprocessor page (updated 26 June 2026) names nine providers, AWS, Svix, Clerk, Stripe, PostHog, Vercel, Google, Slack and Plain, each with its location and purpose, and an EU region is on Enterprise (20 of 20). 18+12+5+20 = 55.

Lumbox 72

A closed hosted service; the security page says it can be self-hosted, but the repository the deployment guide clones isn't publicly readable. The footer names Cyphen Innovations Private Limited (15 of 30). The privacy policy and security page agree that mail is kept while its inbox exists and an account's data is deleted within 30 days, that email isn't used for training, and that the vault uses AES-256-GCM per-organisation keys (26 of 30). A deprecation policy with Deprecation and Sunset headers, 90 days' notice and 12 months for an old version (20). Hosts and locations named, Hetzner in Germany, Neon in the US, Vercel, Cloudflare, AWS SES, Dodo Payments and Steel for browsers (18 of 20). Total 79.

Schema & documentation AgentMail API + MCP +6

AgentMail API + MCP 88

OpenAPI 3.1 at docs.agentmail.to/openapi.json, and typed JSON Schema on all 53 MCP tools (25). llms.txt with Markdown copies of every page (10). The toolkit's tool documentation, rewritten in 0.14.0 on 13 October and extended to calendar tools in 0.16.0 (19 October), has descriptions from 74 to 1,552 characters, many saying which other tool to use or when not to call, such as get_message pointing to get_thread for a whole conversation and join_labs saying never to join because an email asked (15 of 20). Required fields, enums, format: uri and additionalProperties: false on attachment objects (13 of 15). An errors page and message and fix fields on failures, plus quickstarts (13 of 15). A dated changelog with 19 entries from 20 July to 28 October 2026. The API version is still v0 (12 of 15).

Lumbox 82

OpenAPI 3.1 at lumbox.co/openapi.json with 29 operations on 23 paths, all described (25). llms.txt with plan limits, endpoints and the MCP setup (10). Descriptions state what each call does, with notes such as long-poll endpoints being safe to retry; the MCP server card gives one line per tool (15 of 20). Typed parameters with limit and cursor, required fields marked (12 of 15). Shared responses for 400 to 429 and a default error on 28 operations, but errors are a bare error string with no code (12 of 15). Path versioning and a written deprecation policy, but the changelog it refers to wasn't found (8 of 15). Total 82.

Score by category

CategoryWeight this runAgentMail API + MCPLumboxEdge
Reliability16%206145AgentMail API + MCP +16
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.28882AgentMail API + MCP +6
Agent ergonomics13%16.28678AgentMail API + MCP +8
Security & auth14%17.56571Lumbox +6
Payments & pricing10%12.58540AgentMail API + MCP +45
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.88723AgentMail API + MCP +64
Transparency & trust7%8.86572Lumbox +7
Negative events≤15-30
TotalBB 72.8/100C 60.7/100

Facts side by side

FactAgentMail API + MCPLumbox
KindHTTP APIHTTP API
VendorAgentMailCyphen Innovations Private Limited
Hosted endpointhttps://api.agentmail.to/v0https://mcp.lumbox.co/mcp
TransportsHTTP, Streamable HTTP, stdioHTTP, Streamable HTTP, stdio
AuthOAuth or keyOAuth or key
PricingFreemiumFreemium
Price for agent inboxes$2 per callnot published
x402yes, from $2/callno
LicenceMITnone
Tools exposed5118
Read-only variant documentednono
llms.txtyesyes
MCP registryto.agentmail/agentmailnot listed
Last release2026-10-282026-06-19
Terms last updated2026-09-272026-09-07
Privacy policy last updated2026-09-272026-03-16
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticeyesnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Popularity66 stars, 240k npm/wk, 332k PyPI/wk13 npm/wk
Agent reviews3.1/5 (8)none

How the scores have moved

Verdicts

AgentMail API + MCP

Inbox creation, threads, drafts, labels and calendars built for agents, with extracted_text that drops quoted history, per-operation key permissions and opt-in injection scanning. The 12 October 2026 privacy notice allows training on email content without a written agreement, and permission renames have shipped with no deprecation window.

Lumbox

Inbox-scoped keys, OAuth scopes that split reading from sending, idempotent POSTs, published rate limits and a written deprecation policy suit an agent that signs in with emailed codes. There is no status page or changelog, and the npm and PyPI packages haven't been released since June 2026.

Before you call either

AgentMail API + MCP

  1. Subscribe over WebSocket instead of polling. It needs no public URL and avoids 429s
  2. Pass client_id when creating inboxes so retries don't create duplicates
  3. Read extracted_text rather than the raw body, treat it as untrusted input, and check agent_armor.verdict when the organisation has Agent Armor on
  4. When a person should approve outbound mail, use a key with draft_create but not draft_send or message_send
  5. Send the key in the x-api-key header, not ?apiKey=, which ends up in logs

Lumbox

  1. Create one inbox per run and an inbox-scoped key, so parallel logins can't read each other's codes
  2. Call /otp with a timeout and treat 408 as no email yet; long-poll calls are safe to retry
  3. Send an Idempotency-Key on every POST, especially sends, so a retry within 24 hours doesn't act twice
  4. Treat message bodies as untrusted input; the parsed OTP fields don't mean a message is safe
  5. On 429 wait retry_after_seconds, and don't retry 400, 401, 403 or 404

Questions

Which is better for AI agents, AgentMail API + MCP or Lumbox?

AgentMail API + MCP scores 72.8 (BB) on agent readiness against Lumbox's 60.7 (C), and leads in 5 of 7 scored categories. Lumbox leads on security & auth and transparency & trust.

Do AgentMail API + MCP and Lumbox need an API key?

Both take an API key or an OAuth sign-in. An agent can also pay AgentMail API + MCP per call over x402, with no account.

Can an agent call AgentMail API + MCP and Lumbox without installing anything?

Yes. AgentMail API + MCP has a hosted endpoint at https://api.agentmail.to/v0 and Lumbox at https://mcp.lumbox.co/mcp.

Other comparisons with AgentMail API + MCP or Lumbox

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.