2026-03-15 23:05:22 +02:00

Fragment Logo

Fragment API

Python library for the Fragment.com API — purchase Telegram Stars, Premium, and top up TON Ads balance.

PyPI version PyPI downloads Python License Stars CI

Report Bug · Request Feature · Donate TON

Disclaimer: This project is not affiliated with, endorsed by, or in any way officially connected with Fragment or Telegram.


Features

  • 💰 TON Advertisement Topups — Top up Telegram Ads balance (11,000,000,000 TON)
  • 👑 Telegram Premium — Purchase Premium for any user (3, 6, or 12 months)
  • Telegram Stars Purchases — Purchase Stars for any Telegram user (501,000,000 Stars)
  • 🔐 Multi-wallet support — V4R2 and V5R1 wallet contract versions
  • Async-first — Built on httpx and asyncio

📦 Installation

pip install pyfragment

Requires Python 3.12+.


🚀 Quick Start

import asyncio
from pyfragment import FragmentClient

client = FragmentClient(
    seed="word1 word2 ... word24",
    api_key="YOUR_TONAPI_KEY",
    cookies={
        "stel_ssid": "...",
        "stel_dt": "...",
        "stel_token": "...",
        "stel_ton_token": "...",
    },
)

async def main():
    # Purchase 6 months of Telegram Premium
    result = await client.purchase_premium("@username", months=6)
    print(result.transaction_id)

    # Purchase 500 Stars
    result = await client.purchase_stars("@username", amount=500)
    print(result.transaction_id)

    # Top up 10 TON to Ads balance
    result = await client.topup_ton("@username", amount=10)
    print(result.transaction_id)

asyncio.run(main())

See the examples/ folder for ready-to-run scripts.


🔧 Configuration

FragmentClient parameters

Parameter Type Required Default Description
seed str 24-word TON wallet mnemonic phrase
api_key str Tonapi key from tonconsole.com
cookies dict | str Fragment session cookies (dict or JSON string)
wallet_version str "V5R1" Wallet contract version: "V4R2" or "V5R1"

Methods

Usernames can be passed with or without @.

Method Returns Description Limits
purchase_premium(username, months, show_sender=True) PremiumResult Purchase Telegram Premium months: 3, 6, or 12
purchase_stars(username, amount, show_sender=True) StarsResult Purchase Telegram Stars amount: 501,000,000
topup_ton(username, amount, show_sender=True) AdsTopupResult Top up Telegram Ads balance amount: 11,000,000,000
get_wallet() WalletInfo Get wallet address, state, balance

⚙️ Getting Required Credentials

🍪 Fragment.com Cookies

Prerequisites: Log in to fragment.com, connect your TON wallet.

  1. Install Cookie Editor
  2. Open fragment.com while logged in
  3. Click the extension → ExportHeader String
  4. Extract these four fields:
{
  "stel_ssid": "...",
  "stel_dt": "...",
  "stel_token": "...",
  "stel_ton_token": "..."
}

⚠️ Cookies expire. Refresh them if you start getting FragmentPageError or auth errors.

🔑 Tonapi Key

  1. Go to tonconsole.com
  2. Register and generate a new API key
  3. Pass it as api_key to FragmentClient

🌱 Wallet Seed Phrase

If you don't have a TON wallet, create one in Tonkeeper.
Go to Settings → Backup → copy the 24 words.

⚠️ Never share your seed phrase. Store it offline.

🔐 Wallet Version

Version Use when
V5R1 Default — Tonkeeper / MyTonWallet (wallets created after 2024)
V4R2 Older Tonkeeper or hardware wallets

🗂️ Error Handling

All exceptions inherit from FragmentError — see pyfragment/types/exceptions.py for the full list.

from pyfragment import FragmentClient, UserNotFoundError, ConfigurationError, WalletError

try:
    result = await client.purchase_stars("@unknown", amount=100)
except UserNotFoundError:
    print("User not found on Fragment")
except WalletError as e:
    print(f"Wallet issue: {e}")
except ConfigurationError as e:
    print(f"Bad params: {e}")

Made with ❤️ by @bohd4nx

Star this repo if you found it useful!

S
Description
Async Python client for the Fragment API — a unified toolkit to manage Telegram assets: purchase Stars and Premium, top up TON and Ads balances, run giveaways, manage anonymous numbers, and explore the marketplace for usernames, numbers, and gifts.
Readme MIT
259 KiB
Languages
Python 100%