Telegram¶
Quick Start¶
Get your agent up and running in four steps.
Step 1 — Create a Telegram Bot¶
Before creating an agent, set up a bot in Telegram:
- Open Telegram and find @BotFather
- Send the
/newbotcommand - Enter a bot name — it will be displayed in chats (e.g. "MyCompany Support")
- Enter a bot username — a unique identifier ending with
bot(e.g.mycompany_support_bot) - Copy the bot token sent by BotFather
Detailed Guide
Creating a bot takes just a couple of minutes. For a step-by-step guide with screenshots, check out our article.
Step 2 — Create an Agent in Cavuer¶
An agent is an AI profile that manages your channels. You can connect multiple channels to a single agent.
- Sign up at console.cavuer.com
-
On the main page, click "Create Agent", enter a name and select a language model
-
After creating the agent, open its settings and add an instruction — rules for how the agent should interact with customers
Instruction Example
You are a polite support assistant for "MyCompany".
Always address the customer by name.
Only answer questions related to our services.
If you don't know the answer — suggest contacting a live operator.
Choosing a Language Model
Each model differs in speed, accuracy, and cost. For most support tasks, GPT-4 and Claude Haiku models work great. For complex analytical tasks, consider GPT-5, Claude Sonnet, or Claude Opus models.
Step 3 — Connect the Telegram Channel¶
-
On the main page, click "Add Channel" next to your agent and select Telegram
-
Fill in the field:
Field Description Bot Token The token obtained from BotFather in Step 1 -
Click "Create"
Step 4 — Enable the Channel and Start¶
Done! Your AI agent is created, configured, and running. All incoming messages to your Telegram bot will be processed automatically based on the agent's instructions and knowledge base.
Advanced Setup¶
Agent Settings in Cavuer Dashboard¶
Configure the agent's behavior at console.cavuer.com:
| Parameter | Description |
|---|---|
| Name | Rename the agent and channel for easier management in the Cavuer dashboard |
| Model | Select a language model |
| Instruction | Agent behavior rules: greeting, company description, operator transfer conditions, language settings. Specify whether the agent should respond in the customer's language or only in a specific language |
| Response Delay | Delay before responding. Useful when customers send multiple messages in a row — the agent will wait and process all messages together |
| Process Images | Allows the agent to view and analyze image content |
| Process Audio | Enables the agent to recognize voice messages |
| Knowledge Base | Upload .txt, .pdf, or .docx files and connect them in the agent settings |
Full parameter reference: Agent and Channel Setup
Knowledge Base Setup¶
The Knowledge Base is the agent's source of information. There are two ways to populate it:
1. Uploading individual articles. Go to the Knowledge Base section → click "Upload Data" → on the "File" or "Text" tab, enter a title and add the content (.pdf, .docx, .txt files or plain text).
2. Creating collections. Click "Collections" → "Add Collection" and choose a source type: Website, Yandex Wiki, or Confluence. Cavuer will automatically crawl and index the pages.
Then, in the agent settings under Knowledge Base, click "Link Data", select the required articles or collections, and save your changes. For more details, see Knowledge Base.
Knowledge Base Best Practices
- Break up large texts — create smaller articles for specific topics (e.g. separate articles for "Delivery", "Payment", "Returns")
- When editing, use the "Replace Source" function — simple text edits don't update the search index
- Remove unnecessary pages from collections — this improves the accuracy of relevant information retrieval
- Link to documentation — upload a brief summary and include a link to the full version
Usage in Groups¶
The agent can work not only in direct messages but also in Telegram group chats. Learn more: Usage in Groups.
Need Help?
If you have any questions about setting up your agent or connecting a Telegram channel, please contact our support team.



