The Summand MCP server implements the Model Context Protocol and gives any MCP-aware client read access to your Summand datasets and their semantic layer. It’s hosted by Summand atDocumentation Index
Fetch the complete documentation index at: https://docs.summand.com/llms.txt
Use this file to discover all available pages before exploring further.
https://mcp.summand.com/mcp and uses the streamable-HTTP transport with OAuth 2.1 authentication. There’s nothing to install — point your client at the URL, sign in, and start asking questions.
Connect a client
Wire up claude.ai, ChatGPT, or any other OAuth-aware MCP client.
What the server exposes
The server registers seven tools. All of them are read-only — the MCP server cannot mutate datasets, run pipelines, or change settings.| Tool | Purpose |
|---|---|
list_datasets | Discover what datasets the caller can see. |
describe_dataset | Full metadata for a dataset, including which semantic components are available. |
get_columns | Column names and statistics. |
get_semantic_data | Specific semantic components: EBM graphs, feature metadata, embeddings, etc. |
get_surprises | ML-generated surprise findings with explanations. |
get_report_template | The dataset’s current PDF report template. |
compare_schemas | Side-by-side column comparison between two datasets. |
Authentication model
Access is scoped to whatever the caller already has in Summand — owned datasets plus anything explicitly shared with them. There is no way to access a dataset over MCP that you couldn’t access in the dashboard.
- RFC 9728 Protected Resource Metadata — clients auto-discover the authorization server on
401. - RFC 7591 Dynamic Client Registration — no pre-registration required for compliant clients.
What’s next
Connect a client
Step-by-step setup for claude.ai and ChatGPT.
Tools reference
Parameters and example prompts for every tool.