Skip to main content
Glama
uth-mail-mcp

uth-mail-mcp

by fotispetrou

create_draft

Write and save an email draft without sending it, including replies, recipients, and attachments; returns the draft UID for later review or sending.

Instructions

Write a message and save it in Drafts. Nothing is sent.

To reply, give reply_to_uid (and reply_folder if the original is not in the inbox): recipient, 'Re:' subject, threading headers and the quoted original are filled in, and reply_all adds the other recipients as Cc. attachments are paths of local files. Returns the draft's uid for send_draft; the user can also open and edit the draft in webmail.

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

Input Schema

NameRequiredDescriptionDefault
ccNo
toNo
bccNo
bodyYes
subjectNo
reply_allNo
attachmentsNo
reply_folderNoinbox
reply_to_uidNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.5/5.0
Behavior5/5

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

Annotations already declare non-readOnly, non-destructive and non-idempotent, so the bar is lower; the description still adds substantial behavior beyond them — that no message is transmitted, exactly which fields are auto-derived on a reply (recipient, 'Re:' subject, threading headers, quoted original), that reply_all adds the other recipients to Cc, and that the resulting draft is editable in webmail.

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 highest-value fact ('Nothing is sent') in a single short sentence, then a scoped paragraph covering the reply flow. Every sentence carries information, though the reply paragraph is dense with four distinct behaviors packed into one block.

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 9-parameter mutation tool with no output schema and no annotation-level detail, the description covers the send-safety distinction, the reply workflow, attachment semantics, and the return value (draft uid). Remaining gaps like error behavior for an invalid reply_to_uid are minor relative to the complexity.

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 0%, so the description carries the full burden, and it handles the non-obvious parameters well: reply_to_uid triggers recipient/Re: subject/threading-header/quoting behavior, reply_folder handles non-inbox originals, reply_all adds Cc, and attachments are described as local file paths. The self-evident to/cc/bcc/subject/body params are left implicit, which is acceptable because their names are unambiguous.

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 ('Write a message and save it in Drafts') and immediately distinguishes the operation from sending with 'Nothing is sent.' Combined with the sibling list, an agent can tell create_draft apart from send_draft without inspecting either 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?

The 'Nothing is sent' clause establishes the core when-to-use condition and the mention of 'Returns the draft's uid for send_draft' points to the natural follow-up tool. It stops short of an explicit either/or routing statement (e.g. 'use this instead of send when the user wants to review first'), so it is strong context rather than full alternative guidance.

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