Skip to main content
Glama

mb-mastering

by com.mathieuberthet
OverviewDiscussionsAdmin

prepare_quote_email

RÉSERVÉ AU STUDIO (token d'administration requis). Prépare un brouillon — ou envoie — l'email d'un devis ou d'une facture PDF à un client, depuis contact@mathieuberthet.com. Par défaut (mode « draft »), l'email est déposé dans les Brouillons du webmail OVH pour relecture avant envoi ; en mode « send », il est envoyé immédiatement puis archivé dans Envoyés. Le PDF est récupéré par le serveur directement dans Qonto (qonto_type + qonto_number) — ou fourni en base64 pour un document externe.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
ccNoAdresses en copie (string CSV ou array).
modeNodraft (défaut) : brouillon dans le webmail OVH. send : envoi immédiat.
subjectNoSujet de l'email. Défaut : « MB Mastering — Devis/Facture n° … ».
languageNoLangue du wrapper email : 'fr' (défaut, template français avec « Bonjour » + « Bien cordialement, Mathieu Berthet »), 'en' (template anglais avec « Hi » + « Best regards, Jean-Claude assistant to Mathieu Berthet »), ou 'raw' (aucun wrapper — le body_message est envoyé tel quel, pour un contrôle total). Utiliser 'en' pour clients anglophones (SlapFunk, Scott Anderson, etc.) — ne JAMAIS envoyer un body EN avec language 'fr'.
qonto_idNoId Qonto du document (alternative à qonto_number).
auth_tokenNoToken d'administration (si le client MCP ne permet pas le header Authorization).
pdf_base64NoPDF externe encodé en base64 — seulement si le document n'est pas dans Qonto.
qonto_typeNoRECOMMANDÉ : le serveur va chercher le PDF dans Qonto lui-même (aucun base64 à transporter). quote = devis, invoice = facture.
referencesNoChaîne References (Message-IDs séparés par espaces). Auto = in_reply_to si absent.
client_nameNoNom du client (salutation et en-tête To).
in_reply_toNoMessage-ID du mail auquel on répond (avec ou sans chevrons), pour threading IMAP/Gmail.
body_messageNoParagraphe personnalisé inséré dans le corps par défaut.
client_emailYesAdresse email du client destinataire.
pdf_filenameNoNom du fichier joint. Défaut : Devis-<n>.pdf / Facture-<n>.pdf.
qonto_numberNoNuméro Qonto du document (D-2026-003, F-2026-001). Défaut : document_number.
document_typeYesType de document joint.
document_numberYesNuméro du document, ex. 202608-2.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / language
      Added value: +{
+  "description": "Langue du wrapper email : 'fr' (défaut, template français avec « Bonjour » + « Bien cordialement, Mathieu Berthet »), 'en' (template anglais avec « Hi » + « Best regards, Jean-Claude assistant to Mathieu Berthet »), ou 'raw' (aucun wrapper — le body_message est envoyé tel quel, pour un contrôle total). Utiliser 'en' pour clients anglophones (SlapFunk, Scott Anderson, etc.) — ne JAMAIS envoyer un body EN avec language 'fr'.",
+  "enum": [
+    "fr",
+    "en",
+    "raw"
+  ],
+  "type": "string"
+}
  2. Changed5 schema fields changed
    • changedInput schema / properties / pdf_base64 / description
      Previous value: -"Contenu du PDF du document, encodé en base64."New value: +"PDF externe encodé en base64 — seulement si le document n'est pas dans Qonto."
    • addedInput schema / properties / qonto_id
      Added value: +{
+  "description": "Id Qonto du document (alternative à qonto_number).",
+  "type": "string"
+}
    • addedInput schema / properties / qonto_number
      Added value: +{
+  "description": "Numéro Qonto du document (D-2026-003, F-2026-001). Défaut : document_number.",
+  "type": "string"
+}
    • addedInput schema / properties / qonto_type
      Added value: +{
+  "description": "RECOMMANDÉ : le serveur va chercher le PDF dans Qonto lui-même (aucun base64 à transporter). quote = devis, invoice = facture.",
+  "enum": [
+    "quote",
+    "invoice"
+  ],
+  "type": "string"
+}
    • changedInput schema / required
      Previous value: -[
-  "document_type",
-  "document_number",
-  "client_email",
-  "pdf_base64"
-]New value: +[
+  "document_type",
+  "document_number",
+  "client_email"
+]
  3. Changed3 schema fields changed
    • addedInput schema / properties / cc
      Added value: +{
+  "description": "Adresses en copie (string CSV ou array).",
+  "type": [
+    "string",
+    "array"
+  ]
+}
    • addedInput schema / properties / in_reply_to
      Added value: +{
+  "description": "Message-ID du mail auquel on répond (avec ou sans chevrons), pour threading IMAP/Gmail.",
+  "type": "string"
+}
    • addedInput schema / properties / references
      Added value: +{
+  "description": "Chaîne References (Message-IDs séparés par espaces). Auto = in_reply_to si absent.",
+  "type": "string"
+}
  4. Added

TDQS

A3.6/5.0
Behavior4/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It transparently covers the admin token requirement, default draft behavior, immediate send with archiving, PDF retrieval from Qonto or base64, and language template warnings (e.g., never send an EN body with language 'fr'). It does not discuss error cases or idempotency, but the core behaviors are well disclosed.

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?

The description is a single compact paragraph of about four sentences. It front-loads the admin restriction and core action, then explains modes and PDF source. Every sentence contributes; it's appropriately sized and easy to scan. Slight improvement could come from breaking into bullets, but it remains concise and well-structured.

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

Completeness3/5

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

With 17 parameters, no annotations, and no output schema, the description gives a good overview but lacks some workflow context. It doesn't mention expected return values, error conditions, or dependencies with sibling Qonto tools, leaving an agent to infer some integration steps. It is adequate for the main flow but not complete for complex usage.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description adds no substantial parameter-level meaning beyond the schema—it only restates the PDF retrieval options and mode behavior that the schema already documents. No hidden parameter semantics are introduced.

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?

The description clearly states the tool's function: preparing or sending an email with a quote/invoice PDF to a client from a specific address. It specifies the verb (prépare/envoie), resource (email d'un devis/facture), and includes distinguishing context (admin token, studio-only). This makes the purpose unambiguous and distinct from sibling email tools.

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

Usage Guidelines2/5

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

The description gives no explicit guidance on when to use this tool versus alternatives like send_quote_request. It explains the draft/send modes and admin requirement, but does not state selection criteria, exclusions, or prerequisites relative to sibling tools. An agent is left to infer the appropriate context.

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

All mb-mastering toolsInspect
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