Give your team quick answers from Confluence without hunting through pages. A chat message starts a guided lookup that pulls the right page content and turns it into a clear reply for support use. It is great for help desks and internal teams who rely on Confluence for policies, guides, and FAQs.

Here is how it works. A chat message triggers the flow and sets global page IDs to search. The system calls the Confluence API to find the page by ID, then requests the page body in storage format. The content is converted from HTML to clean markdown for better reading. An AI agent with chat memory uses this context to draft the answer and sends it back to chat and to Telegram for fast sharing. The design includes item splitting, schema mapping, and memory so replies stay on topic.

You will need a Confluence API token, an OpenAI API key, and a Telegram bot token. Expect faster support replies and less time spent reading long pages. Teams can reuse the same knowledge with less copy and paste, which reduces mistakes. Common uses include standard operating procedures, product FAQs, and internal setup steps that staff ask about often.

• Chat trigger listens for a new message and starts the lookup flow
• Global variables store one or more Confluence page IDs for quick access
• Confluence search by ID uses the REST API and CQL to locate the page
• Page body is requested in storage format for complete and consistent content
• HTML to Markdown conversion makes long pages easy to read and quote
• An AI agent with memory keeps the conversation context across messages
• Replies are returned to the chat and also sent to Telegram for fast sharing
• Split Out and schema mapping organize items so each page is processed cleanly

• Reduce manual reading time from 10 minutes to 1 minute per question
• Streamline support response prep by about 60 percent using chat memory
• Improve answer accuracy by pulling content from the official page source
• Handle up to 3 times more chat requests without extra staff
• Connect Confluence knowledge and Telegram delivery in one flow

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You will need accounts with Atlassian Confluence, OpenAI and Telegram. See the Tools Required section above for links to create accounts with these services.
3. Create a Confluence API token in your Atlassian account. Build a Basic Auth header by encoding your email and token in Base64 as documented by Atlassian.
4. In n8n, open each Confluence HTTP Request node. In the credential field, click Create new credential, choose HTTP Header Auth, add a header named Authorization with the value Basic followed by your Base64 string. Add Content-Type with value application/json.
5. Replace the example domain with your own site in both Confluence request URLs. Use your actual wiki subdomain and confirm the path matches the v2 pages endpoint and the search endpoint.
6. Open the language model node. In the credential field, click Create new credential, select OpenAI, and paste your OpenAI API key. Save and test the connection.
7. Create a Telegram bot using BotFather and copy the bot token. In the Send Telegram Message node, create a new Telegram credential and paste the token. Set the chat ID for your target chat or user.
8. Open the Globals node and enter your real Confluence page IDs. Keep names clear so you know which page each ID represents.
9. Test the Search By ID node. Execute the node and check for a 200 response and a single result. If you see 404, verify the page ID and your permission to view the page.
10. Run a full test from the chat panel. Ask a question that the selected page can answer. Confirm the reply includes content from that page and that a Telegram message was sent.
11. If you get 401 on Confluence calls, confirm the Authorization header and Base64 value. If the markdown looks messy, check the HTML to Markdown node input path matches the storage body field.
12. Adjust the AI Agent prompt and memory window to control tone and answer length. Save and re test until the reply style fits your support team needs.