- DATA
- LISTS
- SEARCH
- HUMAN
Short answer
How can I get experiment details using my AI?
Your AI uses Neptune.ai tools to fetch specific project data and run attributes. You can pull full JSON objects or target specific metadata fields for any run in your workspace. This allows you to analyze hyperparameters and metrics through a chat interface.
Where your data goes.
Data outcomes
The AI processes your requests into these specific results.
DATA
Raw JSON objects
The AI pulls the full metadata for a specific run. You get the complete attribute set for deep inspection.
LISTS
Project inventories
The AI scans your workspace to find relevant projects. You get a clean list of available experiment containers.
SEARCH
Attribute filtering
The AI looks up specific tags or hyperparameters. You get the exact values tied to a specific run or project.
HUMAN
Manual verification
The AI identifies runs that don't match your query. You decide which experiments to investigate further in the Neptune UI.
The workflow
What your AI does when you ask for metadata.
The AI moves through your Neptune workspace to find the exact data you need.
Locate projects
The AI finds the correct workspace containers to start the search.
list_projectsAccess project data
The AI enters a specific project to see its structure and contents.
get_projectExtract attributes
The AI pulls the specific metadata or hyperparameters you asked for.
get_attributesFormat results
The AI organizes the retrieved JSON or attribute values into a readable response.
get_attributes
Try it
Copy these to start.
Use these prompts to begin interacting with your experiment data.
Starting points
Swap out the project names and run IDs with your actual Neptune.ai identifiers.
Neptune.ai Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_HI0D1zHVFTWAvCQH47Rd6lplgg0GBJXVQZ0fMTHm/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 — Neptune.ai capabilities are ready to use.
{
"mcpServers": {
"neptuneai-ml-experiment-tracking-mcp": {
"url": "https://edge.vinkius.com/vk_preview_HI0D1zHVFTWAvCQH47Rd6lplgg0GBJXVQZ0fMTHm/mcp"
}
}
}List all my projects in Neptune.ai.
Get the attributes for the project named 'image-classifier-v2'.
Show me the metadata for the run with ID 'abc-123'.
What are the specific attributes stored in my 'training-runs' project?
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Neptune.ai once, then ask.
Just provide your credentials through the secure connection link. Once connected, your AI client can immediately start querying your experiments.
Connect Neptune.ai to your AIFAQ
How the Neptune.ai MCP behaves
- 01
Can my AI change experiment parameters?
No. The AI can only read and retrieve metadata; it cannot modify your runs or projects.
- 02
Can the AI delete a project?
No. This MCP is read-only for metadata and attributes. Deletions must be done manually in Neptune.ai.
- 03
How does the AI find my projects?
It uses the list_projects tool to scan your workspace for available project names.
- 04
Can I get the full JSON for a run?
Yes. The AI can pull the detailed attribute sets for any specific run you identify.
- 05
Does the AI see my actual training data?
No. The AI only accesses the metadata and attributes you have configured in Neptune.ai.
More questions about Neptune.ai? The Connector page answers them. See everything the Neptune.ai Connector can do
Connect Neptune.ai to Claude, Cursor, ChatGPT & more
