Skip to main content
Glama

create_draft

Save a plain-text email draft in Mail with optional attachments without sending it; specify the sender when multiple accounts exist.

Instructions

Save a plain text draft in Mail, with files attached when 'attachments' is given. Nothing is sent. With more than one Mail account, 'from' must say which one sends. Ask the user when it is not clear. If Mail is not running, applesauce opens it hidden in the background and leaves it open.

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": {
      "subject": "<subject>",
      "body": "<body>"
    },
    "name": "create_draft"
  }
}

Input Schema

NameRequiredDescriptionDefault
ccNoCc addresses. Each must contain @.
toNoTo addresses. Each must contain @.
bccNoBcc addresses. Each must contain @.
bodyYesPlain text body.
fromNoSender: an account name, or one address of an account (from list_accounts).
subjectYesSubject.
attachmentsNoAbsolute paths of files to attach. At most 10 files, 25 MB for each file and 25 MB in all. Folders, app bundles and files in secret folders (~/.ssh, ~/Library/Keychains and similar) are refused.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.3/5.0
Behavior4/5

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

Annotations already declare non-read-only, non-destructive, closed-world behavior; the description adds meaningful context beyond that: no message is transmitted, and if Mail is not running it is launched hidden and left running, a persistent app-state side effect. It does not say where the draft lands (e.g., Drafts folder) or whether it returns an identifier.

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?

Four short sentences, front-loaded with the core action and the crucial 'nothing is sent' guarantee. Every sentence carries information, though the Mail-launch sentence could be tightened slightly.

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

Completeness4/5

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

For a 7-parameter mutation with full schema coverage and annotations, the description covers purpose, safety, multi-account prerequisites, and app side effects. With no output schema, it would benefit from stating what is returned (e.g., draft identifier) or where the draft is stored, but the gap is minor.

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 baseline is 3; the description still adds interpretation for 'from' by explaining it identifies the sending account when more than one exists and should be resolved with the user. Attachment limits are repeated from the schema rather than extended.

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 ('Save a plain text draft in Mail'), scope conditions (attachments when given), and the key non-effect ('Nothing is sent'), which implicitly separates it from the sibling send_draft. An agent can identify the operation without opening the schema.

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

Usage Guidelines4/5

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

Gives clear contextual guidance: drafts are created rather than sent, and with multiple accounts 'from' must name the sender, with an instruction to ask the user when ambiguous. It stops short of explicitly routing to send_draft for actual delivery, so the when-not path is only implied by 'Nothing is sent.'

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