- DEFINITIONS
- ATTRIBUTES
- SEARCH
- HUMAN
Short answer
How can I find Collibra metadata using my AI?
Your AI uses the Collibra MCP to find assets and pull their specific attributes. You get the exact column definitions you need by asking questions instead of navigating the Collibra dashboard. Here is how it works.
Where the data lands
The outcomes of your request
Your AI handles the searching and fetching so you don't have to.
DEFINITIONS
Direct column meanings
The AI finds the specific column you asked about. You get the text definition right in your chat window.
ATTRIBUTES
Asset property lists
Your agent pulls the full set of attributes for a given asset. This gives you the technical context you need immediately.
SEARCH
Asset discovery
The AI scans Collibra to find the right technical assets. You get a list of matching items to choose from.
HUMAN
Manual verification
You review the metadata the AI pulls. You decide if the definition is the one you actually need for your work.
The workflow
What your AI does when the request arrives.
The AI translates your natural language questions into specific Collibra data calls.
Find the asset
The AI looks through your Collibra environment to find the specific technical asset or column you mentioned.
search_assetsRetrieve attributes
Once the asset is identified, the AI pulls the specific metadata and properties tied to that object.
get_assetFilter details
The AI narrows down the retrieved data to show only the relevant definitions or attributes you asked for.
get_assetPresent findings
The AI presents the final metadata in a clear format within your chat interface.
search_assets
Try it
Copy these to start.
Use these prompts to pull data from Collibra.
Starting points
Replace the bracketed text with your actual Collibra asset names or column IDs.
Collibra Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_fjUuXgIQOi6EM8hgjqwHGSQAawj0vEtDPSUJgRgQ/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 — Collibra capabilities are ready to use.
{
"mcpServers": {
"collibra-mcp": {
"url": "https://edge.vinkius.com/vk_preview_fjUuXgIQOi6EM8hgjqwHGSQAawj0vEtDPSUJgRgQ/mcp"
}
}
}What is the definition for the [column_name] column in Collibra?
Get all attributes for the [asset_name] asset.
Search Collibra for assets related to [data_domain].
Show me the metadata for the [table_name] table.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Collibra once, then ask.
Connect your account via the link below. Your credentials stay encrypted, and you can start asking questions immediately.
Connect Collibra to your AIFAQ
How this task behaves
- 01
Can the AI edit metadata in Collibra?
No. The AI can only read and retrieve existing metadata. It cannot change or delete anything in Collibra.
- 02
Can I use this with Claude or Cursor?
Yes. Any MCP-compatible client will work once you have connected the Collibra MCP.
- 03
Does the AI create new assets?
No. The AI is limited to searching for and getting details on assets that already exist.
- 04
Can the AI delete column definitions?
No. The tools provided are read-only. The AI has no capability to delete data.
- 05
Do I need to provide my Collibra password to the AI?
No. You connect through Vinkius, which manages the connection securely. Your AI client never sees your credentials.
More questions about Collibra? The Connector page answers them. See everything the Collibra Connector can do
Connect Collibra to Claude, Cursor, ChatGPT & more
