Skip to main content
Glama
combor

Baryon MCP

save_draft

Destructive

Create a new Proton Mail draft or replace an existing one by UID, with plain text or HTML bodies, attachments, and reply threading headers for saved emails.

Instructions

Create a complete Proton Mail draft, or replace an existing draft when uid and uidvalidity are provided. Supports plain text, an optional HTML alternative, and bounded attachments supplied either inline as base64 (content_base64) or as a path inside the server's attachment directory that it reads at save time (content_path). To answer one message, save_reply_draft derives its recipients and threading for you; to forward one with its attachments, use save_forward_draft. To reply inside a thread here, read the message with get_email and set in_reply_to to its message_id, and references to its references followed by its message_id; when it reports no references, use its in_reply_to in their place so earlier ancestry survives. A replacement keeps the previous draft's Message-ID, plus whichever of its In-Reply-To and References the call omits; passing an empty array for either removes it, detaching the draft from its thread. Updating appends the replacement before removing the old UID; inspect warning if cleanup was incomplete.

Input Schema

NameRequiredDescriptionDefault
ccNoCc recipient addresses
toNoTo recipient addresses
bccNoBcc recipient addresses retained in the saved draft
uidNoexisting draft UID to replace; requires uidvalidity
fromYessender address; must be one of the Proton account addresses exposed by Bridge, which list_sender_identities reports
subjectNo
html_bodyNooptional HTML alternative, up to 50000 characters
text_bodyNoplain-text body, up to 50000 characters
referencesNoconversation chain: the parent email's references followed by its message_id; set together with in_reply_to so clients thread the reply; when replacing a draft, omit to keep the existing header or pass an empty array to remove it
attachmentsNoregular file attachments; at most 100 and 25 MB decoded in total
in_reply_toNoMessage-IDs this draft replies to, normally just the message_id of the email being answered; angle brackets optional; when replacing a draft, omit to keep the existing header or pass an empty array to remove it
uidvalidityNoDrafts UIDVALIDITY accompanying uid

Output Schema

NameRequiredDescriptionDefault
uidYesUID of the newly saved draft
folderYes
warningNo
uidvalidityYesDrafts folder generation accompanying uid
replaced_uidNo
previous_draft_removedYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.5.0
    • changedInput schema / properties / attachments / items / properties / content_path / description
      Previous value: -"absolute path to a regular file on the server's machine, read when the draft is saved; each attachment takes exactly one of content_base64 or content_path; not available on Windows"New value: +"path to a regular file on the server's machine, read when the draft is saved: relative to the server's attachment directory, or absolute and inside it. Each attachment takes exactly one of content_base64 or content_path; not available on Windows"
    • changedInput schema / properties / from / description
      Previous value: -"sender address; must be one of the Proton account addresses exposed by Bridge"New value: +"sender address; must be one of the Proton account addresses exposed by Bridge, which list_sender_identities reports"
  2. First observedv0.4.1

TDQS

A4.8/5.0
Behavior5/5

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

Annotations declare the mutation profile (destructive, non-idempotent), and the description adds behavior beyond them: a replacement keeps the previous Message-ID and the omitted In-Reply-To/References headers, an empty array removes a header and detaches the thread, and updating appends before removing the old UID with a possible cleanup warning. The attachment path is disclosed as being read server-side at save time.

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 create/replace distinction, then split cleanly into alternatives, threading, and replacement behavior. Dense and mostly waste-free, though the threading sentence about references fallback is a long clause chain that takes effort to unpack.

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 annotations carrying the safety profile and an output schema covering return values, the description supplies the missing pieces an agent needs: source-mode exclusivity for attachments, threading headers, and exact replacement/header-retention semantics. Nothing material is omitted.

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 92%, so the schema documents most parameters. The description nonetheless adds cross-parameter semantics that the schema does not: content_base64 vs content_path being mutually exclusive source modes, and the omit-vs-empty-array rule for references and in_reply_to during a replacement.

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 ('Create a complete Proton Mail draft, or replace an existing draft when uid and uidvalidity are provided'), making the create-vs-replace duality explicit. It names the sibling tools (save_reply_draft, save_forward_draft) and the case each covers, so an agent can route correctly without opening any schema.

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?

Explicit alternatives for the two reply/forward workflows, plus a concrete recipe for replying inside a thread (get_email, set in_reply_to to message_id, references to references + message_id, fall back to in_reply_to when no references). This is when-to-use and how-to-do-it guidance that leaves almost nothing to inference.

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