- INGESTION
- INDEXING
- REVIEW
- USAGE
Short answer
How do I feed my own data into a chatbot?
You give your agent a URL or a document, and it uses ChatFly to add that information to a bot's knowledge base. This makes your specific data available for the agent to reference during chats. You'll see the specific steps below.
The results.
Where your data ends up.
Your agent moves data from your files into a structured format the bot can use.
INGESTION
Data is added
The AI takes the content from your provided links or files. It converts these sources into a format the ChatFly bot can read.
INDEXING
Knowledge is ready
The ingested information is indexed within the bot. Your agent can now pull specific facts from these sources during a conversation.
REVIEW
Human oversight
The AI handles the technical upload, but you decide if the data is correct. You remain in control of the knowledge quality.
USAGE
Bot is active
The bot is ready to answer questions based on the new data. You can test its accuracy immediately in your AI client.
The workflow
What your AI does when the data arrives.
Your agent manages the transfer of information from your sources to the ChatFly bot.
Identify sources
Your agent checks the existing data to see what is already part of the knowledge base.
list_data_sourcesUpload content
The AI pushes the content from your documents or URLs into the bot.
upload_data_sourceVerify ingestion
The agent lists the current sources to confirm the new data was successfully added.
list_data_sourcesFinalize training
The AI completes the process, making the new information available for the bot to use.
upload_data_source
Try it
Copy these to start.
Use these prompts to begin the training process.
Starting points
These are opening lines. Swap the bracketed text for your actual file names or URLs.
ChatFly Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_YeceOYq0Sj0dHYZqDFIY8IgiIyBo9f7LRr4xsK2v/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — ChatFly capabilities are ready to use.
{
"mcpServers": {
"chatfly-alternative-mcp": {
"url": "https://edge.vinkius.com/vk_preview_YeceOYq0Sj0dHYZqDFIY8IgiIyBo9f7LRr4xsK2v/mcp"
}
}
}Upload this PDF to the ChatFly knowledge base: [manual.pdf]
Train the bot using the content from this URL: [https://example.com/docs]
List all the data sources currently in my ChatFly bot.
Add this text file to my knowledge base: [notes.txt]
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect ChatFly once, then ask.
Connect once to give your agent access to ChatFly tools. Your login stays encrypted, and you can find the full walkthrough on the Connector page.
Connect ChatFly to your AIFAQ
How this task behaves.
- 01
Can the AI delete data from my knowledge base?
No. The tools only allow for uploading and listing data sources. You must handle deletions manually in ChatFly.
- 02
Does the AI update the bot if a website changes?
No. The AI only ingests data when you tell it to upload a specific source. It doesn't monitor for live updates.
- 03
Can I see a list of what I've already uploaded?
Yes. You can ask your agent to list all existing data sources.
- 04
Can the AI edit the content of my files?
No. The AI can only upload the files as they are. It cannot change the text inside your documents.
- 05
Can the AI create new bots for me?
No. The tools are limited to managing data sources within existing bots.
More questions about ChatFly? See everything the ChatFly Connector can do
Connect ChatFly to Claude, Cursor, ChatGPT & more
