Busymail

Queue an email for approval

send_email

Queue an email for the account owner to approve. THIS DOES NOT SEND. The message is held in Busymail and goes out only if the owner opens their approval queue and approves it, which they may refuse; it is discarded if they do not decide within a week. Nothing reaches the recipients in the meantime. Never tell the user their message has been sent: say it is waiting for their approval in Busymail. Pass in_reply_to_conversation_id to reply inside an existing thread, which also decides the account it is sent from; otherwise it is a new message from the owner's sending account. Files can be attached by passing their bytes as base64; the owner sees every one of them, named and sized, before deciding.

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": {
      "to": "<to>",
      "subject": "<subject>",
      "body": "<body>"
    },
    "name": "send_email"
  }
}

Input Schema

Name	Required	Description
`cc`	No	Cc recipients, comma-separated. Optional.
`to`	Yes	Recipients, comma-separated. Either bare addresses or 'Name <addr@example.com>'.
`bcc`	No	Bcc recipients, comma-separated. Optional.
`body`	Yes	The message, as plain text. No HTML.
`from`	No	Which of the user's own connected accounts to send from, as its address. Must be one of theirs - it selects an account, it does not set an arbitrary sender. Omit to use their first sending account. A reply always goes from the account its thread is on, so leave this out when replying.
`subject`	Yes	The subject line.
`attachments`	No	Files to attach, each with its bytes inline as base64. Optional. Both the total and what the sending account's own mail server accepts are checked now, when this is called, and a message over either is refused here with the real figures.
`in_reply_to_conversation_id`	No	The UUID of the conversation to reply into, from list_feed, get_cluster or search. Omit for a new message.

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 3, 2026

TDQS

A4.5/5.0

Behavior5/5

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

No annotations exist, so the description carries the full behavioral burden and does so thoroughly: the message is held, the owner must actively approve, they may refuse, it is discarded after a week, and nothing reaches recipients in the interim. It also discloses that attachment size and server limits are checked at call time and refused with real figures, which is exactly the kind of timing behavior an agent could not infer.

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 single most important fact (it does not send), then layered with consequences and parameter behavior. Sentences are dense and each carries information, though the capitalized prohibition and the closing attachments sentence make it longer than strictly minimal.

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 an 8-parameter mutation tool with no annotations and no output schema, the description covers lifecycle, refusal, expiry, reply semantics, and attachment validation. The remaining gap is the return value: it never says what a successful call yields, which matters given that cancel_queued_send and list_queued_sends presumably need some identifier.

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?

With 100% schema coverage the baseline is 3, but the description adds real meaning beyond the schema: in_reply_to_conversation_id also determines the sending account, and attachments are reviewed by name and size by the owner before approval. These are consequences of parameter choices, not restatements of the schema text.

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 ('Queue an email') and immediately disambiguates the most likely confusion with a capitalized 'THIS DOES NOT SEND'. Any agent can tell this apart from a real send operation and from siblings like cancel_queued_send or list_queued_sends 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 strong conditional guidance: pass in_reply_to_conversation_id to reply in a thread, otherwise it's a new message. It also prescribes how to report the outcome to the user ('Never tell the user their message has been sent'), which is genuinely needed usage guidance. It stops short of naming sibling tools or stating when to prefer this over other routes, so not a 5.

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

All Busymail tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?