Skip to main content
Glama
gmail-mcp
by gchallen

create_draft

Create Gmail drafts for new messages or threaded replies, with recipients, HTML or plain text, and local file attachments; returns IDs and a Gmail link.

Instructions

Create a draft. With replyToMessageId it becomes a reply in that thread (subject, In-Reply-To and References set; the original is quoted below the new text unless quote is false); recipients are still yours to give, or use reply with asDraft for Gmail's default recipients. Attachments are local file paths. Returns the draft ID, message ID, thread ID and a Gmail link.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

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

Input Schema

NameRequiredDescriptionDefault
ccNo
toNo
bccNo
bodyNoPlain-text body. One line per paragraph, no hard wrapping; Gmail wraps on display and links bare URLs. Markdown does NOT render: for bullets or other formatting supply htmlBody instead.
quoteNoQuote the replied-to message below the new text. Default true.
subjectNoTitle Case. Defaults to "Re: <original>" when replying.
htmlBodyNoHTML body, used when the message needs formatting such as <ul><li> lists. When given, "body" becomes the plain-text alternative (derived from the HTML if omitted). Do not include the signature; it is appended.
signatureNoSet false to send without the configured signature. By default the signature (from config.json, GMAIL_SIGNATURE, or the Gmail settings signature) is appended once; a copy typed into the body is de-duplicated.
attachmentsNoFiles to attach, as local paths (preferred: any size up to Gmail's 25 MB total) or base64 objects.
replyToMessageIdNoMessage ID to reply to; threads the draft and quotes the original.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv2026.9.2

TDQS

A4.3/5.0
Behavior4/5

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

Annotations declare a non-readonly, open-world, non-destructive mutation; the description adds genuinely non-obvious behavior: subject/In-Reply-To/References are auto-set, the original is quoted unless quote is false, and a configured signature is appended and de-duplicated. It also discloses the return payload (draft ID, message ID, thread ID, Gmail link) and that attachments are local file paths. Missing only operational context like auth or size failure handling, so a 4 rather than 5.

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?

Purpose and the reply behavior are front-loaded, and every sentence carries information that is not in the structured fields. The parenthetical and semicolon-heavy clauses make it dense to parse, but there is no filler to cut.

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?

With 10 optional parameters and no output schema, the description does the important work: explains the only parameter with cross-field effects and enumerates the return values. Remaining gaps are edge cases (invalid paths, quota failures) that are arguably outside a tool description's scope.

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 70%, and the description adds meaning the schema does not: that replyToMessageId does not populate recipients ("recipients are still yours to give"), and that it drives subject, threading headers and quoting. Attachments-as-path is reinforced. It leaves the cc/bcc/body versus htmlBody interplay to the schema, which already documents it well.

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?

Opens with a specific verb+resource ("Create a draft") and immediately differentiates itself from the reply flow by explaining that replyToMessageId turns the draft into a threaded reply, while "reply with asDraft" is named as the alternative for default recipients. An agent can distinguish this from send_draft, update_draft or reply 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 Guidelines4/5

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

Gives clear conditional guidance for the reply case ("With replyToMessageId it becomes a reply in that thread") and names a concrete alternative ("use reply with asDraft"). It does not state exclusions such as when to prefer update_draft over creating a new draft, so it stops short of full when/when-not coverage.

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