Skip to main content
Glama

Save a reply as a draft

draft_reply

Compose a reply to a message and save it in Drafts WITHOUT sending it. Use this whenever the user wants to review or edit a reply before it goes - never draft_email, which composes a standalone message: a draft made that way carries a "Re:" subject but no quoted original and none of the headers that attach it to the thread, so it looks right and arrives as a new conversation. This produces exactly what reply_email would have sent - the original quoted below the new text, correct recipients, correct threading - and puts it in the Drafts folder, where it opens in Outlook or webmail to be edited, sent or deleted. The original is NOT marked as answered, because it has not been answered yet; that happens if and when the user sends it. Nothing is sent and nothing reaches anybody.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
ccNoExtra people to copy in, on top of whoever the original message resolves to - the same thing as typing an address into the Cc line of a reply. Someone already being written to in To is not repeated here.
allNoReply to everyone. The sender and the original To recipients go in To; the original Cc stays in Cc; your own addresses are removed from both. Defaults to false.
uidYesThe UID of the message to reply to, from list_emails.
bodyYesPlain text of the reply - your words only. The quoted original is added below them; do not paste it in yourself.
fromNoSend as one of the other addresses on this mailbox instead of its default. Use it only when the user names an address to send from. It must be one already added to the mailbox; anything else is refused with the list of addresses that would have worked. The mail server still decides whether it will carry the address, and may refuse it at send time.
htmlNoOptional HTML body. Send real markup, used exactly as given and never escaped or unescaped by us: `<p>Hi</p>`, never `&lt;p&gt;Hi&lt;/p&gt;`. Escaped markup is delivered as visible tags and nothing will warn you. Leave it out unless you have real markup to send: the message still goes out as HTML, built from `text`, carrying the mailbox owner's HTML signature. Send it only for real markup of your own words: the quoted original is added to the HTML part either way, and is never something you paste in yourself.
mailboxNoIMAP folder the message is in. Defaults to INBOX.
attachmentsNoFiles to attach. FOUR WAYS IN, AND ONLY ONE OF THEM COSTS YOU ANYTHING. For a file already in this mailbox, call read_email on the message holding it and pass the `ref` from its `attachments` list as `fileRef`. For a file on the web, pass its https:// link as `url` and this server fetches it. For a file on the person's own computer, make a link with create_upload_link and pass its `uploadId` once the file is up - one entry attaches everything that arrived there. Those three work at any size, because the bytes never pass through this conversation. `content` is the last resort and is only for something you generated yourself in this conversation that exists nowhere else: it costs roughly 450,000 tokens per megabyte, so keep it under about 50 KB. All attachments TOGETHER must come to under 10 MB - a whole-message limit, not an allowance per file, so three 4 MB files are refused however they arrived.
plainTextOnlyNoSend with no HTML part at all. Leave this out unless the user has asked for plain text or the recipient is a list that rejects HTML - it drops the mailbox owner's HTML signature and their formatting. Do not set it to be cautious; HTML is the normal format and what their own mail client sends.
confirmReplyToNoOnly needed when this tool refuses because the message carries a Reply-To pointing somewhere it did not come from. Read the refusal, TELL THE USER where the reply is actually going, then pass those addresses here to proceed.
requestReadReceiptNoAsk the recipient's mail program to confirm when the message is opened, exactly as Outlook's "request a read receipt" tick box does. Only set it when the user asks for it. TELL THEM WHAT IT REALLY DOES before promising anything: it is a request their mail program may ignore and that many clients ask them to approve, so most messages never produce one however promptly they are read - consumer Gmail never sends one, Workspace only if an administrator enabled it, Apple Mail only behind a setting that ships off. Use check_receipts afterwards to see whether one came back, and never treat silence there as evidence the message was unread. There is no delivery-receipt option: that needs an SMTP feature none of the mail providers this connects to offer.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. First observed

TDQS

A4.7/5.0
Behavior5/5

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

Beyond the annotations (readOnlyHint: false, openWorldHint: true, etc.), the description discloses critical behavioral traits: nothing is sent, the original is NOT marked as answered, the draft appears in the Drafts folder with quoted original and correct threading, and it opens in Outlook or webmail for later editing, sending, or deletion. This adds substantial behavioral context that annotations alone do not convey.

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

Conciseness5/5

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

The description is four sentences, each earning its place: purpose, when-to-use with sibling contrast, exact output behavior, and explicit non-effects. It is dense but not bloated, front-loads the core purpose, and contains zero redundant phrasing.

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?

Given the tool's complexity (11 parameters, rich schema) and the absence of an output schema, the description covers the essential contextual gaps: what the draft looks like, where it goes, what happens to the original, and that nothing is sent. No output schema means return-value explanation isn't required; the behavioral outcomes are fully described.

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

Parameters3/5

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

Schema description coverage is 100%, so the baseline is 3. The description reinforces that the quoted original is added below the body and that recipients/threading match reply_email, but it does not add new parameter-specific semantics beyond what the schema already provides in detail. It is neither harmful nor a significant value-add over the schema.

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?

The first sentence states the exact verb, resource, and outcome: 'Compose a reply to a message and save it in Drafts WITHOUT sending it.' It explicitly distinguishes itself from draft_email, clarifying the difference between a threaded reply draft and a standalone draft. This is unambiguous and fully differentiates the tool from its nearest sibling.

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?

The description explicitly says when to use this tool ('Use this whenever the user wants to review or edit a reply before it goes') and explicitly warns against using draft_email instead, explaining the technical consequence of that wrong choice. It also relates this tool to reply_email, telling the agent that this produces exactly what reply_email would have sent, which makes the usage decision effortless.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4/5.0
Disambiguation4/5

Most tools target clearly distinct actions, but there are several close pairs that require careful reading: check_bounces vs check_receipts, list_emails vs search_emails, and the send/reply/forward vs draft_email/draft_reply/draft_forward families. The descriptions are strong enough that a careful agent can disambiguate, but the boundaries are not always obvious from the tool names alone.

Naming Consistency5/5

Every tool uses a consistent snake_case verb_noun convention, such as check_bounces, list_mailboxes, send_email, and update_draft. Related operations use parallel forms like mark_read/mark_unread and flag_email/unflag_email, making the naming predictable and coherent.

Tool Count2/5

With 28 tools, this exceeds the 25+ threshold where a tool set starts to feel too heavy. Email is a broad domain and many tools are individually justified, but the surface could be consolidated, especially the draft_forward/draft_reply/draft_email and send_email/reply_email/forward_email variants.

Completeness4/5

The core email lifecycle is thoroughly covered: send, read, search, move, delete, drafts, replies, forwards, folders, flags, attachments, upload links, contacts, identities, deliverability, bounces, and receipts. The main gap is that permanent deletion is referenced as a separate operation but no such tool is actually provided, and contact/identity management is effectively read-only.

Resources