No install? Use cloud.anythingmcp.com directly.
Sign in, install the Telegram Bot in one click, paste the credentials, mint an MCP API key — done. No Docker, no git clone, no local server to run.
TL;DR
Send Telegram messages, photos, documents, audio, location and polls from any AI agent via the Telegram Bot API. 14 tools, bot-token auth. Outbound + lightweight read of recent updates (no long-polling).
💡 インストール不要? cloud.anythingmcp.com を直接利用してください。 サインインし、Connectors → Telegram Bot をクリック、認証情報を貼り付け、MCP API キーを発行 — 完了です。Docker も
git cloneもローカルサーバーも不要。
Telegram Bot + Gemini
Send Telegram messages, photos, documents, audio, location and polls from any AI agent via the Telegram Bot API. 14 tools, bot-token auth. Outbound + lightweight read of recent updates (no long-polling).
前提条件
完全なセットアップ手順はコネクタ自体に組み込まれています (ストアでコネクタを選択すると表示)。必要な環境変数:
TELEGRAM_BOT_TOKEN
ステップ 1 — 認証情報を取得
telegram.org/bots/api).
Setup:
- In Telegram, open chat with @BotFather →
/newbot→ pick a display name and a username ending inbot. - BotFather replies with an HTTP API token like
123456789:ABCDEFghijklmnopqrstuvwxyz. Treat as a secret — anyone with it can send as your bot. - Set
TELEGRAM_BOT_TOKENto the full token.
Authentication model (Telegram-specific): the bot token goes in the URL path, NOT in a header. The base URL is https://api.telegram.org/bot${TELEGRAM_BOT_TOKEN} and every endpoint is appended as /methodName. The adapter substitutes the token via the {{TELEGRAM_BOT_TOKEN}} env-var template in baseUrl.
…(continued in the in-app connector instructions)
ステップ 2 — アダプターをインストール
git clone https://github.com/HelpCode-ai/anythingmcp.git
cd anythingmcp && docker compose up -d
ステップ 3 — Gemini にコネクタを追加
Gemini CLI は ~/.gemini/settings.json (Windows は %APPDATA%\gemini\settings.json) から MCP サーバを読み込みます。追加:
{
"mcpServers": {
"anythingmcp": {
"httpUrl": "https://cloud.anythingmcp.com/mcp",
"headers": { "Authorization": "Bearer YOUR_MCP_API_KEY" }
}
}
}
- AnythingMCP の Profile → MCP API Keys → New Key で MCP API key を取得。
- ファイルを保存し
geminiを再起動。 - Gemini CLI で
/mcpを実行 —Telegram Botが利用可能として表示されます。 - Vertex AI Studio: 同じ Bearer ヘッダーで
https://cloud.anythingmcp.com/mcpをリクエストのtools配列に渡してください。
利用可能なツール
| Tool | What it does |
|---|---|
telegram_bot_get_me | Return the bot's own user info: id, username, first_name, can_join_groups, can_read_all_group_messages, supports_inline_queries |
telegram_bot_send_message | Send a text message to a chat |
telegram_bot_send_photo | Send a photo (JPG/PNG, ≤5MB via URL) by HTTPS URL or by reusable file_id |
telegram_bot_send_document | Send a document (any type, ≤20MB via URL) by HTTPS URL or file_id |
telegram_bot_send_audio | Send an audio file (MP3, M4A, ≤20MB via URL) |
telegram_bot_send_voice | Send a voice note (OGG-Opus encoded) |
telegram_bot_send_video | Send a video (MP4, H |
telegram_bot_send_location | Send a location point |
telegram_bot_send_poll | Send a poll (regular or quiz) |
telegram_bot_edit_message_text | Edit the text of a previously-sent message (your bot's own messages only) |
telegram_bot_delete_message | Delete a message the bot sent (within ~48 hours for groups, anytime for own DMs) |
telegram_bot_get_chat | Fetch chat info: title, type (private/group/supergroup/channel), photo, description, invite_link, member_count, permissions |
telegram_bot_get_updates | Long-poll recent inbound messages (one-shot — pulls all updates since the last offset+1) |
FAQ
Gemini 1.5 Pro / 2.x は MCP をサポートしますか? はい — Gemini CLI ≥ 0.4 と Vertex AI tools API は Bearer ヘッダー付きの httpUrl MCP コネクタを受け入れます。
次のステップ
このガイドは役に立ちましたか?