{
  "data": {
    "category": {
      "area": "communication",
      "capabilities": [
        "email.inbox",
        "email.send",
        "email.inbound",
        "email.threads",
        "email.domains"
      ],
      "description": "APIs that give an agent its own email address, so it can send, receive replies and carry a thread on over days. Compared on how an inbox is created, how replies reach the agent (webhooks, websockets, polling), threading, custom domains and price per inbox.",
      "json": "https://www.anchorterminal.com/categories/agent-inboxes.json",
      "name": "Agent inbox APIs",
      "slug": "agent-inboxes",
      "test": "",
      "title": "Email inbox APIs for AI agents",
      "toolCount": 7,
      "tools": [
        "agentmail",
        "robotomail",
        "mailslurp",
        "mails-ai",
        "inbound",
        "cherami",
        "inboxapi"
      ],
      "url": "https://www.anchorterminal.com/categories/agent-inboxes"
    },
    "faq": [
      {
        "answer": "AgentMail API + MCP has the highest benchmark score of the 7 ranked email inbox APIs, 74.9 (BB). Robotomail is second with 69.8 (B).",
        "question": "What are the highest-rated email inbox APIs for AI agents?"
      },
      {
        "answer": "1 of the 7 ranked here grade BB or better, the bar for agent-ready on the Anchor benchmark.",
        "question": "How many email inbox APIs are agent-ready?"
      },
      {
        "answer": "AgentMail API + MCP. An agent can pay these per call in USDC with no account.",
        "question": "Which email inbox APIs accept x402 payments?"
      },
      {
        "answer": "By the Anchor benchmark score out of 100, a weighted mean of the scored categories minus deductions for negative events, from public evidence re-checked as vendors change. Listings cannot pay for a place. The latest assessment behind this page is from 9 October 2026.",
        "question": "How is this list ranked?"
      }
    ],
    "howToChoose": [
      {
        "label": "How replies reach the agent",
        "detail": "Check whether replies arrive by webhook, websocket or polling, since polling adds delay and cost to every check, and a webhook needs a public endpoint."
      },
      {
        "label": "Inbox creation and isolation",
        "detail": "Check how an inbox is created and whether each can be isolated from others, since an agent that spawns many inboxes must keep their mail apart."
      },
      {
        "label": "Thread retention and matching",
        "detail": "Check how long threads are retained and whether replies are matched to their original conversation, since an agent resuming after days needs its earlier messages still stored."
      },
      {
        "label": "Prompt injection screening",
        "detail": "Check whether inbound messages are scanned for prompt injection before the agent reads them, since an email body is untrusted text that can carry instructions aimed at the agent."
      }
    ],
    "picks": [
      {
        "also": {
          "name": "Robotomail",
          "slug": "robotomail",
          "why": "B, 69.8/100"
        },
        "name": "AgentMail API + MCP",
        "need": "Highest score overall",
        "slug": "agentmail",
        "why": "BB, 74.9/100 on the benchmark"
      },
      {
        "name": "Inbound",
        "need": "Reliability",
        "slug": "inbound",
        "why": "83/100 on reliability, against 61 for the overall leader"
      },
      {
        "name": "Robotomail",
        "need": "Schema \u0026 documentation",
        "slug": "robotomail",
        "why": "90/100 on schema \u0026 documentation, against 85 for the overall leader"
      },
      {
        "name": "mails.ai Agent Email",
        "need": "Security \u0026 auth",
        "slug": "mails-ai",
        "why": "76/100 on security \u0026 auth, against 62 for the overall leader"
      },
      {
        "name": "Robotomail",
        "need": "Transparency \u0026 trust",
        "slug": "robotomail",
        "why": "73/100 on transparency \u0026 trust, against 68 for the overall leader"
      },
      {
        "name": "AgentMail API + MCP",
        "need": "Paying per call with no account (x402)",
        "slug": "agentmail",
        "why": "accepts x402, $2 a call"
      }
    ],
    "ranked": 7,
    "shortlist": [
      {
        "bestFor": "Agents that need their own address and carry email threads over days, including support, scheduling and outreach agents.",
        "grade": "BB",
        "name": "AgentMail API + MCP",
        "position": 1,
        "price": "$2 / call",
        "score": 74.9,
        "slug": "agentmail",
        "strengths": [
          "Inbox creation, threads, drafts and labels built for agents, with `extracted_text` that drops quoted history",
          "x402 on its own domain and an agent sign-up API, so an agent can start without a person in a dashboard",
          "Free plan with 3 inboxes and 3,000 emails a month, no card"
        ],
        "url": "https://www.anchorterminal.com/tools/agentmail",
        "verdict": "Inbox creation, threads, drafts and labels built for agents, with `extracted_text` that drops quoted history. Email sending was down for 8 hours 7 minutes on 19 August 2026.",
        "weaknesses": [
          "Email sending was down for 8 hours 7 minutes on 19 August 2026",
          "API request limits aren't published as numbers",
          "The hosted MCP server's 38 tools load at once, with filtering only in the stdio bridges"
        ],
        "where": "both",
        "x402": "yes"
      },
      {
        "bestFor": "A developer who wants one mailbox per agent with replies as events, on a small fixed monthly price.",
        "grade": "B",
        "name": "Robotomail",
        "position": 2,
        "price": "$19 / mo",
        "score": 69.8,
        "slug": "robotomail",
        "strengths": [
          "OpenAPI 3.1 at `/openapi.json` with 40 operations, plus `llms.txt` and a Markdown copy of every public page",
          "API keys can be limited to named mailboxes, and the MCP server takes OAuth with separate `mail:read` and `mail:send` scopes",
          "Replies reach the agent by signed webhook, by SSE at `GET /v1/events` with replay, or by polling"
        ],
        "url": "https://www.anchorterminal.com/tools/robotomail",
        "verdict": "A public OpenAPI 3.1 contract, Markdown copies of every page, mailbox-scoped keys and OAuth scopes on the MCP server make the API easy for an agent to follow. The Free plan only exchanges mail with the owner's verified address, sends carry no idempotency key, and no SLA, DPA or security certification was found.",
        "weaknesses": [
          "The Free plan allows 10 sends and 10 receives a month, only with the owner's verified address, and discards other inbound mail",
          "Sends take no idempotency key, and the SDKs don't retry them, so an uncertain send has to be checked by hand",
          "No SLA, DPA, SOC 2 or ISO 27001 statement and no bug bounty were found on the site"
        ],
        "where": "hosted",
        "x402": "no"
      },
      {
        "bestFor": "QA and test agents that read OTP, verification and reset mail, and for supervised support agents where a person approves drafts.",
        "grade": "B",
        "name": "MailSlurp",
        "position": 3,
        "price": "$19.99 / mo",
        "score": 68.4,
        "slug": "mailslurp",
        "strengths": [
          "Five agent roles from `AGENT_READ_ONLY` to `AGENT_SUPPORT`, each limited to chosen inbox IDs, inbox tags or the whole account",
          "`AGENT_DRAFT_ONLY` lets an agent write drafts that a person reviews and sends from a review queue",
          "The hosted MCP server takes OAuth with client registration, so no key is pasted into the client"
        ],
        "url": "https://www.anchorterminal.com/tools/mailslurp",
        "verdict": "Agent access is scoped by role and by inbox, with a draft-only role, a human review queue and a per-agent activity trail. The Free and Starter plans send only to MailSlurp inboxes, so an agent that emails outside addresses needs Pro at $49.99 a month, and the terms describe every availability figure as a goal.",
        "weaknesses": [
          "Free sends only to inboxes in the same account and Starter only within MailSlurp. External sending and custom domains start on Pro at $49.99 a month",
          "The account API key is also accepted as an `?apiKey=` query parameter",
          "The OpenAPI file documents only 2xx responses, and 339 of 844 operations have no description"
        ],
        "where": "hosted",
        "x402": "no"
      },
      {
        "bestFor": "Agents that send transactional mail and read replies, such as support, notification and verification agents, where an injection score on inbound mail is wanted out of the box.",
        "grade": "B",
        "name": "mails.ai Agent Email",
        "position": 4,
        "price": "$20 / mo",
        "score": 66.7,
        "slug": "mails-ai",
        "strengths": [
          "API keys scoped to send, read or manage, bindable to one agent, with expiry and immediate revocation",
          "Every inbound message carries an injection score and a quarantined flag",
          "OpenAPI 3.1 spec with 57 operations, llms.txt, an errors page and documented rate limits"
        ],
        "url": "https://www.anchorterminal.com/tools/mails-ai",
        "verdict": "Per-agent addresses with scoped, agent-bound keys, an OpenAPI spec and a prompt-injection score on every inbound message. The operator is unnamed, the vendor calls itself pre-launch, and the status page shows 98.49 per cent API uptime over 90 days with no incident reports.",
        "weaknesses": [
          "No company or person named; legal entity details only on request",
          "98.490 per cent API uptime over 90 days on status.mails.ai, with no incident reports",
          "MCP tools carry no readOnly or destructive hints, and MCP errors drop the error code"
        ],
        "where": "both",
        "x402": "no"
      },
      {
        "bestFor": "A team that wants agent addresses on its own domain at a low fixed price, with IMAP and SMTP beside the API and an MCP mode that confines an agent to one mailbox.",
        "grade": "B",
        "name": "Inbound",
        "position": 5,
        "price": "$9 / mo",
        "score": 63.7,
        "slug": "inbound",
        "strengths": [
          "Mailbox credentials scope reading to chosen addresses or domains and sending to one identity, and the same password works over REST, IMAP, SMTP and MCP",
          "Public OpenAPI 3.1 document with 54 operations, plus `llms.txt`, `llms-full.txt` and Markdown copies of every docs page",
          "The hosted MCP server takes OAuth with PKCE and dynamic client registration, marks every tool with read-only and destructive hints, and narrows its 50 tools with `?toolsets=`"
        ],
        "url": "https://www.anchorterminal.com/tools/inbound",
        "verdict": "A mailbox password limits an agent to chosen addresses and one sending identity across REST, IMAP, SMTP and MCP, and the platform's source is public under MIT. There is no free plan, sign-up needs a browser, webhooks are verified with a static token and are not retried automatically, and no changelog, DPA or security contact was found.",
        "weaknesses": [
          "No free plan or trial on the pricing page. Plans start at $9 a month and the free plan was retired on 3 December 2025",
          "Webhooks carry a static `X-Webhook-Verification-Token` header, not a signature, and failed deliveries are not retried automatically",
          "Account API keys and OAuth tokens (scope `inbound:account`) grant the whole account. Only mailbox passwords are scoped"
        ],
        "where": "hosted",
        "x402": "no"
      },
      {
        "bestFor": "A single developer or small team giving an agent a free, separate address for low-volume recurring correspondence such as supplier replies, project updates and agent-to-agent handoffs.",
        "grade": "C",
        "name": "Cherami",
        "position": 6,
        "price": "Free",
        "score": 57.6,
        "slug": "cherami",
        "strengths": [
          "OpenAPI 3.1 contract with 39 operations, llms.txt and a Markdown twin of every docs page",
          "Every MCP tool carries readOnly, destructive, idempotent and openWorld hints",
          "Idempotency keys on sends, replies, forwards and creation, plus a per-operation retry table"
        ],
        "url": "https://www.anchorterminal.com/tools/cherami",
        "verdict": "Free @cherami.to inboxes with a typed OpenAPI contract, 39 annotated MCP tools and idempotency keys on every send. It's a two-week-old service run by one person, with no status page, no webhooks and keys that reach every inbox on the account.",
        "weaknesses": [
          "No status page, SLA or public changelog",
          "Every API key and OAuth grant reaches all inboxes on the account, with one full-access scope",
          "No webhooks, so an agent has to poll for new mail"
        ],
        "where": "hosted",
        "x402": "no"
      },
      {
        "bestFor": "A personal mailbox for one coding agent that needs to exchange mail with its owner and a few contacts at no cost.",
        "grade": "C",
        "name": "InboxAPI",
        "position": 7,
        "price": "Free",
        "score": 54.5,
        "slug": "inboxapi",
        "strengths": [
          "First use creates the account and address by proof-of-work, with no signup form, API key or card",
          "Every inbound message carries a trust level, and untrusted bodies and subjects are datamarked with a per-request marker",
          "The CLI hides nine auth and encryption tools from the model and requires `confirm` on `forward_email`"
        ],
        "url": "https://www.anchorterminal.com/tools/inboxapi",
        "verdict": "An account and address are created on first use with proof-of-work and no signup, key or card, and inbound mail is trust-labelled and datamarked against prompt injection. Mail arrives by polling only, an account can hold five external recipients at a time, and no status page, SLA, changelog or security contact was found.",
        "weaknesses": [
          "Mail arrives by polling only. No webhook, websocket or push tool was found in the tool list or docs",
          "An account holds five external recipient slots, and a full slot frees only after five days without use",
          "No status page, SLA, changelog, security.txt or disclosure policy was found on the site or in the repository"
        ],
        "where": "both",
        "x402": "no"
      }
    ],
    "updated": "2026-10-09"
  },
  "kind": "anchor.page",
  "links": {
    "api": "https://www.anchorterminal.com/api/v1/index.json",
    "html": "https://www.anchorterminal.com/best/agent-inboxes/",
    "json": "https://www.anchorterminal.com/best/agent-inboxes/index.json",
    "llms": "https://www.anchorterminal.com/llms.txt",
    "markdown": "https://www.anchorterminal.com/best/agent-inboxes/index.md",
    "slim": "https://www.anchorterminal.com/best/agent-inboxes/index.min.md"
  },
  "markdown": "All 7 ranked email inbox APIs on the Anchor benchmark, with a pick for each need and where each one falls short. Scores come from public evidence, re-checked as vendors change.\n\n- Ranked: 7 · agent-ready (BB or better): 1 · accept x402: 1 · hosted endpoints: 7\n- Full ranked table: https://www.anchorterminal.com/categories/agent-inboxes.md\n- Head-to-head comparisons: https://www.anchorterminal.com/compare/agent-inboxes/index.md (21)\n- Methodology: https://www.anchorterminal.com/benchmark/index.md\n\n## The shortlist\n\n| # | Tool | Grade | Score | Best for | Price | Where |\n| --- | --- | --- | --- | --- | --- | --- |\n| 1 | [AgentMail API + MCP](https://www.anchorterminal.com/tools/agentmail.md) | BB | 74.9 | Agents that need their own address and carry email threads over days, including support, scheduling and outreach agents. | $2 / call | hosted and local |\n| 2 | [Robotomail](https://www.anchorterminal.com/tools/robotomail.md) | B | 69.8 | A developer who wants one mailbox per agent with replies as events, on a small fixed monthly price. | $19 / mo | hosted |\n| 3 | [MailSlurp](https://www.anchorterminal.com/tools/mailslurp.md) | B | 68.4 | QA and test agents that read OTP, verification and reset mail, and for supervised support agents where a person approves drafts. | $19.99 / mo | hosted |\n| 4 | [mails.ai Agent Email](https://www.anchorterminal.com/tools/mails-ai.md) | B | 66.7 | Agents that send transactional mail and read replies, such as support, notification and verification agents, where an injection score on inbound mail is wanted out of the box. | $20 / mo | hosted and local |\n| 5 | [Inbound](https://www.anchorterminal.com/tools/inbound.md) | B | 63.7 | A team that wants agent addresses on its own domain at a low fixed price, with IMAP and SMTP beside the API and an MCP mode that confines an agent to one mailbox. | $9 / mo | hosted |\n| 6 | [Cherami](https://www.anchorterminal.com/tools/cherami.md) | C | 57.6 | A single developer or small team giving an agent a free, separate address for low-volume recurring correspondence such as supplier replies, project updates and agent-to-agent handoffs. | Free | hosted |\n| 7 | [InboxAPI](https://www.anchorterminal.com/tools/inboxapi.md) | C | 54.5 | A personal mailbox for one coding agent that needs to exchange mail with its owner and a few contacts at no cost. | Free | hosted and local |\n\n## Picks by need\n\n- Highest score overall: [AgentMail API + MCP](https://www.anchorterminal.com/tools/agentmail.md), BB, 74.9/100 on the benchmark. Also [Robotomail](https://www.anchorterminal.com/tools/robotomail.md), B, 69.8/100.\n- Reliability: [Inbound](https://www.anchorterminal.com/tools/inbound.md), 83/100 on reliability, against 61 for the overall leader.\n- Schema \u0026 documentation: [Robotomail](https://www.anchorterminal.com/tools/robotomail.md), 90/100 on schema \u0026 documentation, against 85 for the overall leader.\n- Security \u0026 auth: [mails.ai Agent Email](https://www.anchorterminal.com/tools/mails-ai.md), 76/100 on security \u0026 auth, against 62 for the overall leader.\n- Transparency \u0026 trust: [Robotomail](https://www.anchorterminal.com/tools/robotomail.md), 73/100 on transparency \u0026 trust, against 68 for the overall leader.\n- Paying per call with no account (x402): [AgentMail API + MCP](https://www.anchorterminal.com/tools/agentmail.md), accepts x402, $2 a call.\n\n## How to choose\n\n- How replies reach the agent: Check whether replies arrive by webhook, websocket or polling, since polling adds delay and cost to every check, and a webhook needs a public endpoint.\n- Inbox creation and isolation: Check how an inbox is created and whether each can be isolated from others, since an agent that spawns many inboxes must keep their mail apart.\n- Thread retention and matching: Check how long threads are retained and whether replies are matched to their original conversation, since an agent resuming after days needs its earlier messages still stored.\n- Prompt injection screening: Check whether inbound messages are scanned for prompt injection before the agent reads them, since an email body is untrusted text that can carry instructions aimed at the agent.\n\n## Each one in detail\n\n### 1. AgentMail API + MCP, BB 74.9/100\n\nEmail inboxes for agents over REST.\n\n- Verdict: Inbox creation, threads, drafts and labels built for agents, with `extracted_text` that drops quoted history. Email sending was down for 8 hours 7 minutes on 19 August 2026.\n- Choose it for: Agents that need their own address and carry email threads over days, including support, scheduling and outreach agents.\n- Strength: Inbox creation, threads, drafts and labels built for agents, with `extracted_text` that drops quoted history\n- Strength: x402 on its own domain and an agent sign-up API, so an agent can start without a person in a dashboard\n- Strength: Free plan with 3 inboxes and 3,000 emails a month, no card\n- Weakness: Email sending was down for 8 hours 7 minutes on 19 August 2026\n- Weakness: API request limits aren't published as numbers\n- Weakness: The hosted MCP server's 38 tools load at once, with filtering only in the stdio bridges\n- Price: $2 / call · Auth: OAuth or key · x402: yes · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/agentmail.md\n\n### 2. Robotomail, B 69.8/100\n\nRobotomail gives an AI agent its own email address for sending, receiving and replying. Access is by REST API, a hosted MCP server with OAuth, a CLI and SDKs, and replies arrive by webhook, SSE or polling.\n\n- Verdict: A public OpenAPI 3.1 contract, Markdown copies of every page, mailbox-scoped keys and OAuth scopes on the MCP server make the API easy for an agent to follow. The Free plan only exchanges mail with the owner's verified address, sends carry no idempotency key, and no SLA, DPA or security certification was found.\n- Choose it for: A developer who wants one mailbox per agent with replies as events, on a small fixed monthly price.\n- Strength: OpenAPI 3.1 at `/openapi.json` with 40 operations, plus `llms.txt` and a Markdown copy of every public page\n- Strength: API keys can be limited to named mailboxes, and the MCP server takes OAuth with separate `mail:read` and `mail:send` scopes\n- Strength: Replies reach the agent by signed webhook, by SSE at `GET /v1/events` with replay, or by polling\n- Weakness: The Free plan allows 10 sends and 10 receives a month, only with the owner's verified address, and discards other inbound mail\n- Weakness: Sends take no idempotency key, and the SDKs don't retry them, so an uncertain send has to be checked by hand\n- Weakness: No SLA, DPA, SOC 2 or ISO 27001 statement and no bug bounty were found on the site\n- Price: $19 / mo · Auth: OAuth or key · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/robotomail.md\n- Against #1: https://www.anchorterminal.com/compare/agentmail-vs-robotomail.md\n\n### 3. MailSlurp, B 68.4/100\n\nMailSlurp is a hosted email and SMS API from Pettman OÜ of Estonia. It creates real inboxes and phone numbers for tests and for supervised AI agents, reached over REST, SDKs, IMAP and SMTP, and a hosted MCP server.\n\n- Verdict: Agent access is scoped by role and by inbox, with a draft-only role, a human review queue and a per-agent activity trail. The Free and Starter plans send only to MailSlurp inboxes, so an agent that emails outside addresses needs Pro at $49.99 a month, and the terms describe every availability figure as a goal.\n- Choose it for: QA and test agents that read OTP, verification and reset mail, and for supervised support agents where a person approves drafts.\n- Strength: Five agent roles from `AGENT_READ_ONLY` to `AGENT_SUPPORT`, each limited to chosen inbox IDs, inbox tags or the whole account\n- Strength: `AGENT_DRAFT_ONLY` lets an agent write drafts that a person reviews and sends from a review queue\n- Strength: The hosted MCP server takes OAuth with client registration, so no key is pasted into the client\n- Weakness: Free sends only to inboxes in the same account and Starter only within MailSlurp. External sending and custom domains start on Pro at $49.99 a month\n- Weakness: The account API key is also accepted as an `?apiKey=` query parameter\n- Weakness: The OpenAPI file documents only 2xx responses, and 339 of 844 operations have no description\n- Price: $19.99 / mo · Auth: OAuth or key · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/mailslurp.md\n- Against #1: https://www.anchorterminal.com/compare/agentmail-vs-mailslurp.md\n\n### 4. mails.ai Agent Email, B 66.7/100\n\nHosted email API that gives each AI agent its own address to send, receive and thread mail, with a prompt-injection scan on every inbound message. Reached over REST, TypeScript and Python SDKs, and an MCP server.\n\n- Verdict: Per-agent addresses with scoped, agent-bound keys, an OpenAPI spec and a prompt-injection score on every inbound message. The operator is unnamed, the vendor calls itself pre-launch, and the status page shows 98.49 per cent API uptime over 90 days with no incident reports.\n- Choose it for: Agents that send transactional mail and read replies, such as support, notification and verification agents, where an injection score on inbound mail is wanted out of the box.\n- Strength: API keys scoped to send, read or manage, bindable to one agent, with expiry and immediate revocation\n- Strength: Every inbound message carries an injection score and a quarantined flag\n- Strength: OpenAPI 3.1 spec with 57 operations, llms.txt, an errors page and documented rate limits\n- Weakness: No company or person named; legal entity details only on request\n- Weakness: 98.490 per cent API uptime over 90 days on status.mails.ai, with no incident reports\n- Weakness: MCP tools carry no readOnly or destructive hints, and MCP errors drop the error code\n- Price: $20 / mo · Auth: API key · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/mails-ai.md\n- Against #1: https://www.anchorterminal.com/compare/agentmail-vs-mails-ai.md\n\n### 5. Inbound, B 63.7/100\n\nInbound is an email API from Exon Enterprise LLC for sending, receiving and replying on a customer's own domains. Agents use its REST API, webhooks, scoped IMAP and SMTP mailboxes, a hosted MCP server and the `inboundctl` CLI.\n\n- Verdict: A mailbox password limits an agent to chosen addresses and one sending identity across REST, IMAP, SMTP and MCP, and the platform's source is public under MIT. There is no free plan, sign-up needs a browser, webhooks are verified with a static token and are not retried automatically, and no changelog, DPA or security contact was found.\n- Choose it for: A team that wants agent addresses on its own domain at a low fixed price, with IMAP and SMTP beside the API and an MCP mode that confines an agent to one mailbox.\n- Strength: Mailbox credentials scope reading to chosen addresses or domains and sending to one identity, and the same password works over REST, IMAP, SMTP and MCP\n- Strength: Public OpenAPI 3.1 document with 54 operations, plus `llms.txt`, `llms-full.txt` and Markdown copies of every docs page\n- Strength: The hosted MCP server takes OAuth with PKCE and dynamic client registration, marks every tool with read-only and destructive hints, and narrows its 50 tools with `?toolsets=`\n- Weakness: No free plan or trial on the pricing page. Plans start at $9 a month and the free plan was retired on 3 December 2025\n- Weakness: Webhooks carry a static `X-Webhook-Verification-Token` header, not a signature, and failed deliveries are not retried automatically\n- Weakness: Account API keys and OAuth tokens (scope `inbound:account`) grant the whole account. Only mailbox passwords are scoped\n- Price: $9 / mo · Auth: OAuth or key · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/inbound.md\n- Against #1: https://www.anchorterminal.com/compare/agentmail-vs-inbound.md\n\n### 6. Cherami, C 57.6/100\n\nCherami gives an agent free @cherami.to email inboxes for recurring correspondence, reached through a hosted MCP server, an HTTP API with an OpenAPI contract, and TypeScript and Python SDKs. A person approves access in the browser.\n\n- Verdict: Free @cherami.to inboxes with a typed OpenAPI contract, 39 annotated MCP tools and idempotency keys on every send. It's a two-week-old service run by one person, with no status page, no webhooks and keys that reach every inbox on the account.\n- Choose it for: A single developer or small team giving an agent a free, separate address for low-volume recurring correspondence such as supplier replies, project updates and agent-to-agent handoffs.\n- Strength: OpenAPI 3.1 contract with 39 operations, llms.txt and a Markdown twin of every docs page\n- Strength: Every MCP tool carries readOnly, destructive, idempotent and openWorld hints\n- Strength: Idempotency keys on sends, replies, forwards and creation, plus a per-operation retry table\n- Weakness: No status page, SLA or public changelog\n- Weakness: Every API key and OAuth grant reaches all inboxes on the account, with one full-access scope\n- Weakness: No webhooks, so an agent has to poll for new mail\n- Price: Free · Auth: OAuth or key · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/cherami.md\n- Against #1: https://www.anchorterminal.com/compare/agentmail-vs-cherami.md\n\n### 7. InboxAPI, C 54.5/100\n\nInboxAPI gives an AI agent its own email address on a subdomain of inboxapi.ai for sending, receiving, searching, replying and forwarding. Access is through MCP, by a local CLI that bridges stdio to the hosted service.\n\n- Verdict: An account and address are created on first use with proof-of-work and no signup, key or card, and inbound mail is trust-labelled and datamarked against prompt injection. Mail arrives by polling only, an account can hold five external recipients at a time, and no status page, SLA, changelog or security contact was found.\n- Choose it for: A personal mailbox for one coding agent that needs to exchange mail with its owner and a few contacts at no cost.\n- Strength: First use creates the account and address by proof-of-work, with no signup form, API key or card\n- Strength: Every inbound message carries a trust level, and untrusted bodies and subjects are datamarked with a per-request marker\n- Strength: The CLI hides nine auth and encryption tools from the model and requires `confirm` on `forward_email`\n- Weakness: Mail arrives by polling only. No webhook, websocket or push tool was found in the tool list or docs\n- Weakness: An account holds five external recipient slots, and a full slot frees only after five days without use\n- Weakness: No status page, SLA, changelog, security.txt or disclosure policy was found on the site or in the repository\n- Price: Free · Auth: None · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/inboxapi.md\n- Against #1: https://www.anchorterminal.com/compare/agentmail-vs-inboxapi.md\n\n## Head to head\n\n- [AgentMail API + MCP vs Robotomail](https://www.anchorterminal.com/compare/agentmail-vs-robotomail.md)\n- [AgentMail API + MCP vs MailSlurp](https://www.anchorterminal.com/compare/agentmail-vs-mailslurp.md)\n- [AgentMail API + MCP vs mails.ai Agent Email](https://www.anchorterminal.com/compare/agentmail-vs-mails-ai.md)\n- [AgentMail API + MCP vs Inbound](https://www.anchorterminal.com/compare/agentmail-vs-inbound.md)\n- [MailSlurp vs Robotomail](https://www.anchorterminal.com/compare/mailslurp-vs-robotomail.md)\n- [mails.ai Agent Email vs Robotomail](https://www.anchorterminal.com/compare/mails-ai-vs-robotomail.md)\n- [Inbound vs Robotomail](https://www.anchorterminal.com/compare/inbound-vs-robotomail.md)\n- [mails.ai Agent Email vs MailSlurp](https://www.anchorterminal.com/compare/mails-ai-vs-mailslurp.md)\n- [Inbound vs MailSlurp](https://www.anchorterminal.com/compare/inbound-vs-mailslurp.md)\n- [Inbound vs mails.ai Agent Email](https://www.anchorterminal.com/compare/inbound-vs-mails-ai.md)\n\n## Questions\n\n### What are the highest-rated email inbox APIs for AI agents?\n\nAgentMail API + MCP has the highest benchmark score of the 7 ranked email inbox APIs, 74.9 (BB). Robotomail is second with 69.8 (B).\n\n### How many email inbox APIs are agent-ready?\n\n1 of the 7 ranked here grade BB or better, the bar for agent-ready on the Anchor benchmark.\n\n### Which email inbox APIs accept x402 payments?\n\nAgentMail API + MCP. An agent can pay these per call in USDC with no account.\n\n### How is this list ranked?\n\nBy the Anchor benchmark score out of 100, a weighted mean of the scored categories minus deductions for negative events, from public evidence re-checked as vendors change. Listings cannot pay for a place. The latest assessment behind this page is from 9 October 2026.\n\n## How this list is made\n\nThe order is the Anchor benchmark score, the same number as on each listing. Each listing is graded from public evidence against the benchmark checklist, and the picks are worked out from those grades, prices and facts. No listing pays for its place, and paid audits or listing help never change a score.\n",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-10",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.4",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  },
  "page": {
    "breadcrumbs": [
      {
        "name": "Home",
        "url": "https://www.anchorterminal.com/"
      },
      {
        "name": "Best of",
        "url": "https://www.anchorterminal.com/best/"
      },
      {
        "name": "Agent inbox APIs",
        "url": ""
      }
    ],
    "description": "AgentMail API + MCP (BB), Robotomail (B) and MailSlurp (B) lead the 7 ranked email inbox APIs. Picks by need, strengths, weaknesses and prices from the Anchor benchmark.",
    "facts": [
      "AgentMail API + MCP BB",
      "Robotomail B",
      "MailSlurp B"
    ],
    "h1": "Best email inbox APIs for AI agents",
    "image": "https://www.anchorterminal.com/assets/og/best-agent-inboxes.png",
    "path": "/best/agent-inboxes/",
    "published": "",
    "section": "tools",
    "title": "Best email inbox APIs for AI agents in 2026, ranked | Anchor Terminal",
    "toc": null,
    "updated": "2026-10-09",
    "url": "https://www.anchorterminal.com/best/agent-inboxes/"
  },
  "tokens": {
    "markdown": 4450,
    "slim": 1180
  },
  "version": 1
}