Corpus

What the corpus MCP server does

The corpus MCP server exposes read-only semantic search for a collection of Vedic material. The searchable content includes scripture verses, commentaries, and recorded lectures. Semantic search is useful when a query describes an idea or topic without using the exact wording found in the source material.

Results are intended to help an AI agent locate relevant material within the corpus. Because the service is described as read-only, the available listing supports retrieval only; it does not indicate that agents can add, edit, delete, or annotate content.

How it works

An MCP-compatible client can use the corpus MCP server to submit a search request against the available Vedic sources. The server then searches the indexed verses, commentaries, and lecture content by meaning. The supplied information does not document query parameters, result fields, ranking behavior, pagination, authentication, or whether searches can be limited to one content type.

The listing also does not identify a separate tool name or provide examples of requests and responses. Treat semantic search as the confirmed capability, while considering the exact invocation format a setup detail that must be checked in the server's own documentation.

Tools and capabilities

The confirmed capability is:

• Read-only semantic search over Vedic scripture verses
• Read-only semantic search over commentaries
• Read-only semantic search over recorded lectures
• Retrieval based on semantic relevance rather than only exact phrase matching

No other tools are documented in the supplied material. In particular, there is no support stated for writing content, managing records, uploading sources, or performing general-purpose web search.

Limitations and notes

The available listing contains no README, so installation instructions and runtime configuration cannot be verified. It does not state whether the corpus MCP server runs over standard input/output, HTTP, or another transport. No environment variables, package commands, client configuration examples, license, authentication requirements, or version information are provided.

Client compatibility is also undocumented. An MCP client may require additional configuration, but the supplied material does not name Claude Desktop, Cursor, or any other client specifically. Before adopting the corpus MCP server, confirm how it is started, how a client connects, whether access is public or authenticated, and what response format the search tool returns.

The corpus MCP server should be evaluated as a focused retrieval service for Vedic source material. It is a fit for agents that need relevant passages and teachings; it should not be selected on the assumption that it supports content updates or broader research functions.