Skip to main content
Glama
deenesjakoruzh
SymbolStar

gmail-mcp

by SymbolStar
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

Gmail MCP Server

Gmail MCP Server は、Node.js/TypeScript ベースの MCP stdio サーバーであり、OpenClaw が Model Context Protocol を通じて Gmail のメールを読み取れるようにします。

機能

  • Gmail API OAuth 2.0 認証を使用。権限範囲は読み取り専用：https://www.googleapis.com/auth/gmail.readonly

  • list_emails：受信トレイのメールを一覧表示。maxResults および Gmail の query フィルタをサポート

  • get_email：messageId を通じて個別のメール詳細を読み取り

  • search_emails：Gmail の検索構文を使用してメールを検索

  • list_labels：すべての Gmail ラベル/フォルダを一覧表示

  • MCP stdio トランスポートをサポートし、OpenClaw からの呼び出しが可能

Related MCP server: gmail-mcp

環境要件

  • Node.js 18 以上

  • npm

  • Gmail API にアクセス可能な Google アカウント

インストール

npm install
npm run build

Google Cloud Console での OAuth 認証情報の作成

  1. Google Cloud Console を開きます。

  2. プロジェクトを作成または選択します。

  3. APIs & Services -> Library に移動し、Gmail API を検索して有効にします。

  4. APIs & Services -> OAuth consent screen に移動します。

  5. ユーザータイプを選択します。個人利用の場合は通常 External を選択します。

  6. アプリ名、ユーザーサポートメール、デベロッパー連絡先メールアドレスを入力します。

  7. Scopes ステップで Gmail の読み取り専用権限を追加します：https://www.googleapis.com/auth/gmail.readonly。

  8. アプリがテスト状態の場合は、Test users に自分の Gmail アカウントを追加します。

  9. APIs & Services -> Credentials に移動します。

  10. Create Credentials -> OAuth client ID をクリックします。

  11. Application type で Desktop app を選択します。

  12. 作成後に JSON ファイルをダウンロードします。

credentials.json の配置

設定ディレクトリを作成し、ダウンロードした OAuth JSON を以下のように保存します：

mkdir -p ~/.gmail-mcp
chmod 700 ~/.gmail-mcp
cp /path/to/downloaded/client_secret.json ~/.gmail-mcp/credentials.json
chmod 600 ~/.gmail-mcp/credentials.json

最終的なパスは以下の通りである必要があります：

~/.gmail-mcp/credentials.json

認証フローの実行

npm run auth

スクリプトは以下の処理を行います：

  • 一時的なローカル OAuth コールバックサーバーを起動

  • ブラウザを開いて Google 認証を実行

  • 認証完了後、トークンを ~/.gmail-mcp/token.json に保存

ブラウザが自動的に開かない場合は、ターミナルに表示された認証 URL を手動でブラウザにコピーしてください。

MCP Server の起動

npm run start

start は stdio トランスポートを使用します。通常は OpenClaw が MCP サーバーのサブプロセスとして起動するため、手動で常時実行する必要はありません。

OpenClaw MCP 設定例

コマンドを本プロジェクトのビルド後のエントリポイントに向けます：

{
  "mcpServers": {
    "gmail": {
      "command": "node",
      "args": ["/Volumes/DevDisk/symbol/gmailMCP/dist/src/index.js"]
    }
  }
}

プロジェクトディレクトリ内で npm を使用して起動することも可能です：

{
  "mcpServers": {
    "gmail": {
      "command": "npm",
      "args": ["run", "start"],
      "cwd": "/Volumes/DevDisk/symbol/gmailMCP"
    }
  }
}

Tools パラメータの説明

list_emails

受信トレイのメールを一覧表示します。

{
  "maxResults": 10,
  "query": "from:example@gmail.com newer_than:7d"
}

  • maxResults：オプション、デフォルト 10、最大 50

  • query：オプション、Gmail 検索構文、INBOX 内の検索に限定されます

get_email

個別のメール詳細を読み取ります。

{
  "messageId": "18f..."
}

返されるフィールドには、送信者、受信者、件名、日付、ラベル、本文テキスト、HTML 本文、および添付ファイルのメタデータが含まれます。

search_emails

Gmail のメールを検索します。

{
  "query": "subject:invoice has:attachment newer_than:30d",
  "maxResults": 10
}

  • query：必須、Gmail 検索構文をサポート

  • maxResults：オプション、デフォルト 10、最大 50

list_labels

すべてのラベル/フォルダを一覧表示します。

{}

ファイルの場所

  • OAuth credentials：~/.gmail-mcp/credentials.json

  • OAuth token：~/.gmail-mcp/token.json

  • MCP server エントリポイント：dist/src/index.js

よくある質問

Missing Gmail OAuth credentials

Google OAuth クライアント JSON がダウンロード済みであり、以下に保存されていることを確認してください：

~/.gmail-mcp/credentials.json

Missing Gmail OAuth token

先に以下を実行してください：

npm run auth

access_denied またはアプリが未検証

OAuth consent screen がまだテスト状態の場合は、現在の Gmail アカウントを Test users に追加する必要があります。

invalid_grant

古いトークンを削除してから再認証してください：

rm ~/.gmail-mcp/token.json
npm run auth

Available Tools

4 tools
get_emailA

Read one Gmail message by messageId.

NameRequiredDescriptionDefault
messageIdYesGmail message id.

TDQS

A3.8/5.0
Behavior3/5

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

No annotations are provided, so the description must disclose behavioral traits. It states 'Read,' which implies a read-only operation with no side effects, but does not detail error handling, permissions required, or what the response contains. This is minimally adequate for a simple read tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, well-structured sentence with no unnecessary words. It is front-loaded with the verb and resource, making it efficient for an agent to parse.

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?

Given the tool's simplicity (one parameter, no output schema), the description is largely complete. It could be improved by briefly noting what the tool returns (e.g., message content or metadata), but it still provides enough context for basic 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?

The input schema has 100% description coverage for the single parameter (messageId: 'Gmail message id.'). The description adds no further meaning beyond what the schema already provides, so it meets the baseline of 3.

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 uses the specific verb 'Read' and clearly identifies the resource as 'one Gmail message by messageId.' It distinguishes from siblings like list_emails (list multiple) and search_emails (search), making the tool's purpose unambiguous.

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

Usage Guidelines3/5

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

The description implies that the tool is used when you have a specific messageId, but it does not explicitly state when to use it versus alternatives (e.g., list_emails for browsing, search_emails for filtering). No exclusions or prerequisites are mentioned.

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

list_emailsB

List Gmail inbox messages. Supports maxResults and Gmail query filters.

NameRequiredDescriptionDefault
maxResultsNoMaximum number of emails to return. Default 10, max 50.
queryNoOptional Gmail search query, applied within INBOX.

TDQS

B3.1/5.0
Behavior2/5

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

No annotations provided, so description must disclose behavior. Mentions it lists inbox messages and supports parameters, but no details on pagination, response format, or conditions like invalid queries. For a read operation, minimal disclosure.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, no filler. Efficiently conveys the core purpose and supported features.

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?

For a simple tool with 2 params and no output schema, the description covers basics but lacks usage guidelines and behavioral details, leaving gaps for an AI agent to make informed decisions.

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 coverage is 100% with clear parameter descriptions (maxResults: default, min, max; query: optional within INBOX). Description merely restates that parameters are supported, adding no new semantic meaning beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

Clearly states 'List Gmail inbox messages' with verb and resource, distinguishing from get_email (single email) and list_labels (labels). However, does not explicitly differentiate from search_emails, which may also list emails but with different scope.

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?

No guidance on when to use this tool vs alternatives like search_emails. Only states what it does, leaving the agent to infer usage context.

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

list_labelsA

List all Gmail labels and folders.

NameRequiredDescriptionDefault

No parameters

TDQS

A4.3/5.0
Behavior3/5

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

No annotations are provided, so the description is the sole source. It states the action but lacks details like pagination, data scope, or authentication requirements. For a simple list, this is adequate but minimal.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

A single, front-loaded sentence contains all necessary information without waste.

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?

Given the tool's simplicity (0 parameters, no output schema) and clear sibling differentiation, the description is complete enough for an agent to use correctly.

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?

The input schema has no parameters and 100% coverage. With 0 parameters, the baseline is 4, and the description does not add parameter information beyond the schema.

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 'List all Gmail labels and folders' uses a specific verb and resource, clearly distinguishing from sibling tools that deal with emails (get_email, list_emails, search_emails).

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?

While no explicit when-to-use or alternatives are stated, the sibling context makes it clear this is for labels/folders vs. emails, implying appropriate usage.

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

search_emailsA

Search Gmail messages with Gmail search syntax.

NameRequiredDescriptionDefault
queryYesGmail search query string.
maxResultsNoMaximum number of emails to return. Default 10, max 50.

TDQS

A3.6/5.0
Behavior2/5

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

No annotations provided, so description bears full burden. It only mentions search syntax, but does not disclose output format, error behavior, or whether results include full messages or metadata.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Single sentence with no redundancy. Every word adds meaning.

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

Completeness2/5

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

Lacks details on return format (e.g., fields, pagination) and how it differs from list_emails. No output schema, so description should provide more context.

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 100%, so baseline is 3. Description adds value by specifying query uses 'Gmail search syntax', which is essential for correct invocation.

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?

Description clearly states verb (Search) and resource (Gmail messages) using specific syntax. Distinguishes from sibling tools like get_email (single) and list_labels.

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

Usage Guidelines3/5

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

No explicit guidance on when to use search vs alternatives like list_emails or get_email. The description is minimally adequate but lacks contextual differentiation.

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

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 4 tool updatesv0.0.1
    • First observedget_email
    • First observedlist_emails
    • First observedlist_labels
    • First observedsearch_emails
Deploy Server
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

TDQS

A3.6/5.0

Scored across 4 tools

Disambiguation5/5

Each tool targets a distinct operation: reading a single email by ID, listing inbox messages, searching with Gmail syntax, and listing labels. There is no overlap in functionality.

Naming Consistency5/5

All tool names follow the consistent verb_noun pattern in snake_case: get_email, list_emails, list_labels, search_emails.

Tool Count4/5

With 4 tools, the set is concise and scoped to basic read and list operations. While it covers common use cases, it could be expanded with write operations, but the count is reasonable for a focused subset.

Completeness2/5

The tool surface lacks write operations such as sending, deleting, or modifying emails and labels. Users cannot perform actions beyond reading and listing, which is a significant gap for comprehensive Gmail management.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • Gmail MCP Server

    cablate
    A
    license
    Not graded
    quality
    D
    maintenance
    Provides comprehensive Gmail integration with LLM processing capabilities, allowing users to read, search, filter emails and handle attachments through the Model Context Protocol.
    25 npm
    2
    MIT

  • gmail-mcp

    dedalus-labs
    A
    license
    Not graded
    quality
    D
    maintenance
    A Model Context Protocol server that enables full management of Gmail accounts, including messages, threads, labels, and drafts. It leverages the Gmail API to allow searching, sending, and organizing emails via secure OAuth2 authentication.
    MIT

  • Gmail MCP Server

    Imsharad
    F
    license
    Not graded
    quality
    D
    maintenance
    Provides access to Gmail functionality through the Model Context Protocol, allowing LLMs like Claude to interact with your email.
    2
    -

  • Gmail MCP Server

    TechVest-Global
    F
    license
    Not graded
    quality
    D
    maintenance
    Enables AI agents to manage Gmail through natural language, including search, read, send, label, and draft operations via the Model Context Protocol.
    -