- INDEXED
- GROUNDED
- ORGANIZED
- HUMAN
Short answer
How do I give my AI access to my own files and websites?
You upload your documents or point the agent to a URL to create a semantic index. Once the data is ingested, your agent uses that index to find relevant information for any prompt. You get an agent that actually knows your specific context.
Where every piece of data lands.
The outcomes of your data ingestion.
Once the process finishes, your agent has a structured way to look up information.
INDEXED
Searchable data index
Your files and web pages are converted into a semantic format. The AI can now find meaning in your data, not just keywords.
GROUNDED
Fact-based responses
The agent uses the index to pull specific details. This stops the AI from making things up when you ask about your files.
ORGANIZED
Structured storage
All your information lives in a dedicated bucket. This keeps your knowledge base separate and easy to manage.
HUMAN
Manual review needed
You decide which files or URLs are worth adding. The AI won't decide what's important to index without your input.
The workflow
What your AI does when you provide data.
The agent handles the heavy lifting of organizing and retrieving your information.
Set up storage
The agent prepares a dedicated space to hold your indexed information.
create_bucketAdd your files
You provide local documents which the agent processes into the index.
ingest_documentsAdd web content
The agent crawls specific websites to pull in their text and structure.
ingest_websiteRetrieve answers
When you ask a question, the agent looks through the index to find the right context.
search_content
Try it
Copy these to start.
Use these prompts to trigger the data ingestion and search process.
Starting points
These are starting points. Replace the bracketed text with your actual file names or URLs.
GroundX Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_mAVynUnnXcCNvXWpgEW86XlUhDZSOLIyZ2eNgeMy/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 — GroundX capabilities are ready to use.
{
"mcpServers": {
"groundx-mcp": {
"url": "https://edge.vinkius.com/vk_preview_mAVynUnnXcCNvXWpgEW86XlUhDZSOLIyZ2eNgeMy/mcp"
}
}
}Create a new bucket named 'project-alpha-docs' for my research.
Ingest the documents in my 'legal-contracts' folder into the index.
Add the content from https://example.com/docs to my knowledge base.
Search the index for any mention of the Q3 budget requirements.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect GroundX once, then ask.
Just click the link to connect. Your credentials stay encrypted and the setup is handled on our end.
Connect GroundX to your AIFAQ
How the knowledge base works
- 01
Can the AI delete my original files?
No. The agent only reads your files to create an index; it cannot modify or delete your local files.
- 02
How does the AI find the right information?
It uses the search tool to perform a semantic search against the index you created.
- 03
Can the AI browse the entire internet?
No. It can only ingest specific websites that you explicitly tell it to crawl.
- 04
Does the AI decide which files to index?
No. You must manually trigger the ingestion for specific documents or websites.
- 05
Can I use this with Claude or Cursor?
Yes. Any MCP-compatible client can use this once you connect GroundX.
More questions about GroundX? The Connector page answers them. See everything the GroundX Connector can do
Connect GroundX to Claude, Cursor, ChatGPT & more
