🚀 Telegram RAG-Ready GROUP Scraper

Extract Users, Messages, & Metadata from Communities for AI & Analytics.

[!NOTE] Honesty First: This tool is designed for GROUPS (Public or Private Supergroups). It is NOT for Channels (Broadcasts). You cannot scrape members from a Channel unless you are an Admin. For the best results (RAG, User History, Hidden Members), use this on active Groups.

Are you building an AI agent, researching a community, or generating leads? The Telegram RAG-Ready Scraper is the only tool designed specifically for the AI era. It doesn't just scrape; it structures data so your LLM can understand it instantly.

---

🌟 Why Choose This Actor?

🕵️‍♂️ Uncover "Hidden" Members

Most scrapers only see what's on the member list (often just Admins). We go deeper.

• Smart Discovery: detects active users from chat history, even if they are hidden from the member list.
• Get the Real Crowd: Capture the active community, not just the silent observers.

🧠 Built for RAG (Retrieval Augmented Generation)

Stop wasting tokens cleaning messy data. We provide a rag_text field for every user:

• Contextualized History: A single, chronological text block of a user's recent messages.
• Ready to Embed: Feed this directly into Pinecone, Weaviate, or OpenAI embeddings.

🧹 Clean & Rich Data

• Bot Filtering: We identify bots and exclude their spammy messages from the context, keeping your AI pure.
• Detailed Metadata: Get Group IDs, Creation Dates, and precise User/Bot counts.
• Role Detection: Automatically tags users as admin, moderator, bot, or user.

---

📊 Comprehensive Output

We don't just give you a list. We give you a Knowledge Graph.

{
  "group_info": {
    "name": "Python Developers Community",
    "stats": { "human_count": 1450, "bot_count": 12 },
    "created_at": "2020-05-21T..."
  },
  "users": [
    {
      "id": "999888777",
      "username": "AI_Researcher",
      "role": "user",
      "rag_text": "User @AI_Researcher (ID: 999..) sent 5 messages:\n[2024-12-14...] Has anyone fine-tuned Llama 3 yet?",
      "messages": [ ...full message objects... ]
    }
  ]
}

---

⚡ Use Cases

• AI Agents: Give your bot specific knowledge about who said what in a community.
• Lead Generation: Find active, engaged users discussing specific topics (crypto, coding, marketing).
• Community Analysis: Identify key influencers (Admins/Mods) and filter out noise (Bots).
• Sentiment Analysis: Analyze grouped user history to track sentiment over time.

---

🛠️ How to Use

1. Get API Credentials (First Time Only)

To use the scraper, you need a developer api_id and api_hash from Telegram. It is free and takes 1 minute.

1. Go to my.telegram.org.
2. Log in with your phone number.
3. Click on API development tools.
4. Create a new application (values don't matter much):
   • App title: My Scraper
   • Short name: myscraper
   • Platform: Desktop
5. Copy your App api_id and App api_hash.

2. Generate Session String

You need a Session String to log in securely.

[!IMPORTANT] Generate this LOCALLY. For security, Telegram requires you to verify your login on your own device first.

1. Run the interactive setup script:

   $node generate_session.js
   

2. Enter the api_id and api_hash you got in Step 1.
3. Follow the colorful on-screen prompts to log in.
4. Copy the Session String handling it carefully (it's your secret key!).

3. Configure on Apify

• sessionString: Paste the string you generated above.
• groupUrl: The target group, e.g., https://t.me/example_group.
• maxItems: Set to 0 for UNLIMITED scraping (scrapes everyone).
• maxMessages: Set to 0 for UNLIMITED history context.

---

🔒 Privacy & Safety

• Phone Numbers: We respect Telegram's privacy settings. Phones are only scraped if the user has made them public.
• Passive Scraping: This actor reads data like a normal user. It does not spam or join/leave groups rapidly.

---

Monetize your data. Train your agents. Understand your community. Start scraping today.