From telegram
Add Telegram sticker packs with AI-generated descriptions for each sticker. Use when the user wants to add a sticker pack, list known packs, or remove one.
How this skill is triggered — by the user, by Claude, or both
Slash command
/telegram:stickersThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Downloads sticker packs from Telegram, converts each sticker to a viewable
Downloads sticker packs from Telegram, converts each sticker to a viewable image, and generates a JSON mapping of sticker file_unique_ids to human-readable descriptions. This lets you recognize stickers in future conversations without re-downloading them.
Arguments passed: $ARGUMENTS
~/.claude/channels/telegram/sticker-packs.json:
{
"<pack_name>": {
"title": "Human-readable pack title",
"stickers": {
"<file_unique_id>": "<description of what the sticker depicts>"
}
}
}
Missing file = {}.
~/.claude/channels/telegram/sticker-packs.json (handle missing)./telegram:stickers add <pack_name>.add <pack_name>This is the main workflow. It downloads every sticker in the pack, converts animated ones to PNG, shows each to you for description, and saves the mapping.
sticker-packs.json (create {} if missing).curl -s "https://api.telegram.org/bot${TELEGRAM_BOT_TOKEN}/getStickerSet?name=<pack_name>"
Read the bot token from ~/.claude/channels/telegram/.env.title and the stickers array.getFile + the file URL.
b. If the file is .webm or .tgs, convert to PNG using:
nix-shell -p ffmpeg --run "ffmpeg -i <input> -frames:v 1 -y <output.png>"
c. If the file is .webp, it can be read directly.
d. Read the resulting image file with the Read tool.
e. Write a short description of what the sticker depicts (character,
emotion, action — 5-15 words). Include the emoji if present.
f. Record: stickers[file_unique_id] = description
g. Clean up the downloaded/converted files.sticker-packs.json (pretty-printed, 2-space indent).Important: Process stickers in batches of ~5 to avoid overwhelming context. Read 5 images, describe them, then move on.
remove <pack_name>sticker-packs.json.describe <pack_name>sticker-packs.json.Use ~/.claude/channels/telegram/inbox/ as a temp directory for downloaded
sticker files. Clean up after processing each batch.
The bot token is in ~/.claude/channels/telegram/.env as
TELEGRAM_BOT_TOKEN=<value>. Read it with:
grep TELEGRAM_BOT_TOKEN ~/.claude/channels/telegram/.env | cut -d= -f2
npx claudepluginhub solastrius/claude-plugins --plugin telegramProvides CLI for Telegram Bot API using bot tokens: authenticate, send messages, read chats, get info, manage multi-bot setups. For CI/CD and server-side messaging automation.
Adds Telegram reaction directives (`[react:<emoji>]`) to assistant replies. The bot strips the tags and applies the first valid reaction.
Implements Telegram bots with full Bot API support: BotFather setup, messages, webhooks, inline keyboards, groups, channels. Node.js and Python boilerplates.