Skip to main content
Glama
Sign In
Email Sending MCP

Email Sending MCP

by resend
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

Reply to Inbox Thread Email

reply-to-inbox-thread-email
Destructive

Reply to a specific message in a Resend inbox thread by sending a real email to its author or all recipients, with optional cc/bcc.

Instructions

Purpose: Reply to one message in a Resend inbox thread. This sends a real email the moment you call it, from the inbox's address to the author of the message, or to everyone on it with replyAll, plus anyone you add in cc or bcc. At least one of html or text is required.

NOT for: Passing the message along to someone new without answering it (use forward-inbox-thread-email), or starting a new conversation from the inbox (use send-new-inbox-email).

Returns: The sent email's id.

When to use: User asks to reply to, answer, or follow up on a message in an inbox thread, and has given you the wording. Before using this tool, you MUST double-check with the user that they want to send this reply. Reference the WORDING you are about to send and the SUBJECT of the thread you are answering when double-checking, and warn the user that a reply cannot be undone: a sent email cannot be recalled. You may only use this tool if the user explicitly confirms they want to send the reply after you double-check. You do not choose who the reply answers: by default the reply goes only to the author of the message you answer, at its Reply-To address if it has one. If the inbox sent that message, the reply goes to the people it was sent to. Set replyAll to true when the user wants to reply to everyone on that message. Pass cc or bcc only to add recipients the user named. Mention in the double-check whether you reply to the author only or to everyone, and any cc or bcc recipients you add. A reply can reach at most 50 recipients in total, counted after replyAll adds everyone on the message. Name any attachments in the double-check. Omit subject to use the thread's. Get the emailId from get-inbox-thread.

Input Schema

NameRequiredDescriptionDefault
ccNoEmail address to copy on the reply, or an array of addresses. Not copied from the message you answer. With replyAll, these join the copied recipients, without duplicates.
bccNoEmail address to blind-copy on the reply, or an array of addresses. Not copied from the message you answer, even with replyAll.
htmlNoHTML body of the reply. Provide html, text, or both. One of the two is required.
textNoPlain text body of the reply. Provide html, text, or both. One of the two is required.
emailIdYesID of the message being replied to. This is the "id" get-inbox-thread returns for each message in the thread.
headersNoOptional custom email headers as key/value pairs (e.g. {"X-Entity-Ref-ID": "unique-id"}). Header names cannot be empty, and names and values cannot contain line breaks or null characters. In-Reply-To, References and Reply-To are reserved: Resend sets them, and passing any of them, in any letter case, fails.
inboxIdYesInbox ID, or the inbox email address (e.g. "support@example.com"). Not the Resend receiving address.
subjectNoSubject for the reply. Omit to use the thread's subject, prefixed with "Re:", which is usually what you want. Max 2000 characters. It cannot be only whitespace or contain a line break.
replyAllNoReply to everyone on the message instead of only its author. For a message the inbox received, the reply goes to the author as usual and copies the message's other to recipients and its cc recipients. For a message the inbox sent, it goes to its Reply-To when that is not the inbox's address and copies its to and cc recipients; otherwise it goes to its to recipients and copies its cc recipients. The inbox's own address is left out and duplicates are removed. Defaults to false (author only).
threadIdYesThread ID
attachmentsNoFiles to attach. Each one needs filename plus content or url. An empty file or a blocked file type (such as .exe) is rejected. The content of all files can total at most 25 MB. A url file does not count toward it, because it is fetched when the email is sent. The whole email, encoded, can be at most 40 MB.
idempotencyKeyNoOptional unique key that prevents duplicate sends on retries (sent as the Idempotency-Key header). Use the same key when retrying the same logical email; use a new key for a different email. Must be 1-256 characters.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed5 schema fields changedv2.27.1
    • changedInput schema / properties / attachments / description
      Previous value: -"Files to attach. Each one needs filename plus content or url. An empty file or a blocked file type (such as .exe) is rejected. Up to 10 files. The content of all files can total at most 25 MB. A url file counts toward the 10 files only, because it is fetched when the email is sent. The whole email, encoded, can be at most 40 MB."New value: +"Files to attach. Each one needs filename plus content or url. An empty file or a blocked file type (such as .exe) is rejected. The content of all files can total at most 25 MB. A url file does not count toward it, because it is fetched when the email is sent. The whole email, encoded, can be at most 40 MB."
    • changedInput schema / properties / attachments / items / properties / content / description
      Previous value: -"Base64-encoded file content. Provide content or url."New value: +"Base64-encoded file content. Use `url` instead for files larger than a few hundred kilobytes. Provide content or url."
    • changedInput schema / properties / attachments / items / properties / url / description
      Previous value: -"URL where the file is hosted. Resend fetches it when the email is sent. It must be an http or https URL that does not point to localhost, 127.0.0.1 or 0.0.0.0. Provide content or url."New value: +"URL where the file is hosted. Resend fetches it when the email is sent. It must start with http:// or https:// and cannot start with http://localhost:. A URL Resend cannot fetch fails the email when it is sent, not this call. Provide content or url."
    • removedInput schema / properties / attachments / maxItems
      Removed value: -10
    • changedInput schema / properties / inboxId / description
      Previous value: -"Inbox ID"New value: +"Inbox ID, or the inbox email address (e.g. \"support@example.com\"). Not the Resend receiving address."
  2. Addedv2.26.1

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare destructiveHint=true and openWorldHint=true, and the description reinforces this with materially new context: real email sends immediately, cannot be recalled, a 50-recipient cap counted after replyAll, default routing logic (author vs. everyone), and reserved header behavior. This is substantial disclosure beyond the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Uses labeled sections (Purpose, NOT for, Returns, When to use) that are front-loaded and scannable. It is long, but for an irreversible send operation almost every sentence carries a distinct obligation; the density is justified.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Covers a 12-parameter, nested-object mutation with a stated return value ('the sent email's id') and no output schema, and closes the important gaps: confirmation requirement, recipient limits, default recipient resolution, and attachment constraints. Nothing an agent needs to call it safely is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds meaning the schema does not: omit subject to inherit the thread subject with 'Re:', a reply reaches at most 50 recipients after replyAll expansion, and attachments must be named in the confirmation. These reinforce key constraints rather than merely restating fields.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Reply to one message in a Resend inbox thread') and explicitly distinguishes itself from the two nearest siblings, forward-inbox-thread-email and send-new-inbox-email, with named use cases for each. An agent can route between the three without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit when-to-use, an explicit NOT-for section naming alternatives, prerequisites (get emailId from get-inbox-thread), and a mandatory double-check-with-user workflow. Nothing about selection is left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools