Skip to main content
Glama

send_draft

Destructive

Send a saved Mail draft by its Drafts mailbox ID after confirming recipients and attachments; applesauce rebuilds and sends it, then deletes the original draft.

Instructions

Send a draft that is saved in a Drafts mailbox. Give its id from search_mail (with 'mailbox' Drafts). Mail cannot send a saved draft directly, so applesauce builds a new message from the draft (subject, sender, To, Cc, Bcc, plain text body and attachments), sends it, and then deletes the old draft (Mail moves it to the Trash). Formatting of a rich text draft is lost. A draft with an attachment that cannot be fetched, or with a text above 100000 characters, is not sent. Without 'confirm' true nothing is sent: the result shows the subject, the sender, the recipients and the attachment names, so ask the user first. A message outside a Drafts mailbox is refused. If Mail is not running, applesauce opens it hidden in the background and leaves it open. Sending cannot be undone. create_draft and forward_message do not return the id of the draft they save. Find it with search_mail.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "id": "<id>"
    },
    "name": "send_draft"
  }
}

Input Schema

NameRequiredDescriptionDefault
idYesMessage id of the draft, from search_mail in a Drafts mailbox.
confirmNoMust be true to send.
in_accountNoAccount id of the message (the 'account_id' field of search_mail). Only needed with 'in_mailbox'.
in_mailboxNoMailbox path of the message (the 'mailbox' field of search_mail). Only needed when Full Disk Access is missing. Give 'in_account' too.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.8/5.0
Behavior5/5

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

Goes well beyond the annotations (readOnlyHint=false, destructiveHint=true) by explaining the mechanism: a new message is rebuilt from the draft, sent, and the old draft is deleted to Trash. It discloses failure modes (unfetchable attachment, body over 100000 characters, non-Drafts message) that silently prevent sending, plus side effects like loss of rich text formatting and Mail being launched hidden and left open.

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?

Front-loaded with the action and its precondition, and nearly every sentence carries a distinct fact (dry-run behavior, size limit, irreversibility, background launch). It is dense and delivered as one undifferentiated block, which slightly obscures the most safety-critical points.

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?

With no output schema, the description still explains what the result contains in the non-confirming case and that sending is irreversible and deletes the draft. For a destructive, multi-step tool it leaves nothing an agent needs in order to call it safely.

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 already 100%, so the baseline is 3, but the description adds real meaning beyond it: 'confirm' false acts as a dry run returning subject, sender, recipients and attachment names, and the id must come from search_mail restricted to a Drafts mailbox. It does not further explain the in_account/in_mailbox fallback conditions beyond what the schema already says.

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 ('Send a draft that is saved in a Drafts mailbox') and immediately scopes it to drafts obtained from search_mail. It clearly separates itself from create_draft and forward_message, whose saved drafts lack a returned id, so an agent can route correctly without inspecting other schemas.

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?

Explicitly states the precondition (id from search_mail with 'mailbox' Drafts), the exclusion ('A message outside a Drafts mailbox is refused'), and the required confirmation flow ('Without confirm true nothing is sent ... ask the user first'). It also names the sibling tools whose output must be resolved via search_mail first.

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