> For the complete documentation index, see [llms.txt](https://docs.ovaledge.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ovaledge.com/release8.1/mcp-server/model-context-protocol.md). # Model Context Protocol The Model Context Protocol (MCP) is an open integration standard for secure access to external systems and contextual metadata by AI assistants and intelligent applications. MCP defines a standard framework for connecting large language models (LLMs) and AI-driven workflows with enterprise platforms. The OvalEdge MCP Server exposes governed metadata, lineage information, glossary data, and platform knowledge through MCP-compatible interfaces. AI clients such as Claude Desktop, Cursor, Visual Studio Code extensions, etc., can retrieve metadata directly from OvalEdge while respecting existing access controls. All requests are executed using the authenticated user’s permissions. Metadata visibility is governed by the access controls enforced in the OvalEdge platform. {% hint style="warning" %} The following upgrades must be completed to access OvalEdge MCP server: * **Application Version**: Migration to the 8.1.x release. * **Tech Stack Upgrades**: Upgrade to Elasticsearch 9.2 and Java 21. {% endhint %} ## OvalEdge MCP Tools The OvalEdge MCP Server provides read-only tools to search, explore, and understand governed metadata within the OvalEdge platform. These tools return structured responses that include governance context such as ownership, stewardship, classifications, and data quality indicators. ### Search & Discovery #### **Semantic Catalog Search (search\_catalog\_assets)** Search cataloged assets using natural-language requests and hybrid search techniques that combine keyword relevance with semantic vector matching. Results include governance metadata and contextual asset information to help AI clients efficiently identify relevant data assets. #### **Asset Metadata Retrieval (catalog\_asset\_details)** Retrieves detailed metadata for a specific catalog object, including object properties, schema details, ownership information, classifications, and governance attributes. Assets can be retrieved using either a fully qualified name or an internal object identifier. #### **Glossary Term Lookup (lookup\_glossary\_term)** Accesses the governed business glossary terms and definitions maintained in OvalEdge. The tool connects business terminology with associated physical data assets, related concepts, and governance context for AI client processing. #### **Business Glossary Discovery (search\_glossary\_metadata)** Searches and retrieves Business Glossary metadata using natural-language queries. The tool enables users to discover glossary terms, domains, categories, governance ownership, related terms, and glossary hierarchies while preserving existing access controls and governance policies. Users can explore business definitions, governance stakeholders, and associated glossary content, and navigate directly to glossary terms and domains via application links. #### **Tag Discovery (lookup\_tags)** Retrieves classification tags and metadata labels associated with governed assets. This capability provides AI systems with sensitivity classifications, governance tagging structures, and organizational metadata standards. #### **Metadata Change Analysis (get\_metadata\_changes\_between\_crawls)** Analyzes metadata changes between crawls to help users understand how cataloged assets have evolved over time. The tool identifies additions, deletions, and modifications across schemas, tables, and columns, and provides summarized insights into structural changes and impacted objects. Users can compare changes across selected crawl periods using natural-language queries and access related comparison details for further investigation. #### **Data Story Discovery (lookup\_datastory)** Retrieves Data Story information stored within OvalEdge, including Story Zone details, content sections, metadata, and access information. The tool enables users to search for Data Stories by name or identifier, explore documented business knowledge, verify access permissions, view authorized users and roles, and navigate directly to the Data Story through a provided application link. #### **Lineage Traversal (asset\_lineage)** Explores upstream and downstream lineage relationships for cataloged assets. The OvalEdge MCP Server enables AI clients to trace dependencies across datasets, files, and related systems for data movement and impact analysis workflows. #### **Entity Relationship Discovery (table\_entity\_relationships)** Retrieves structural relationships between database entities, including primary-key and foreign-key mappings. This capability provides relationship information for AI applications to identify dataset connectivity and generate join recommendations for analytical workflows. #### **Column Profiling Information (column\_profile\_statistics)** Accesses profiling statistics for supported table and file assets. Profiling information includes column-level characteristics for governance-aware metadata analysis. #### **Documentation Search (search\_platform\_docs)** The OvalEdge MCP Server includes documentation search capabilities that help users find relevant procedures, how-to guides, and usage information for the OvalEdge application through natural-language queries. #### **Native Source System Access Discovery (get\_source\_system\_access)** Retrieves native access permissions and privilege information directly from supported source systems. The tool enables users to identify who has access to specific objects, determine which objects a user can access, and understand how access was granted through direct permissions, roles, or group memberships. Access information is available across Amazon Redshift, Snowflake, and Tableau through natural language queries, providing visibility into source-system permissions independent of OvalEdge catalog access controls. ### Governance Actions #### **Governance Role Management (update\_governance\_roles)** Manages governance role assignments for supported assets through natural-language interactions. Users can assign, update, or remove roles such as Owner, Steward, Custodian, and other configured governance roles while ensuring compliance with existing permissions, inheritance rules, and governance policies. Role updates can be applied to individual assets or across related child assets when governance inheritance is enabled. #### **Business Glossary Term Creation (create\_terms)** Creates new Business Glossary terms through natural-language interactions. Users can define business terms, descriptions, domains, categories, and subcategories, ensuring that terms are added to the appropriate glossary hierarchy. The tool validates user permissions, identifies potential duplicate terms, and provides guidance when additional information is needed to maintain a consistent and governed Business Glossary. #### **Tag Creation (create\_tags)** Creates and manages governance, business, and classification tags through natural-language interactions. The tool enables users to define tag attributes, organize tags within the appropriate hierarchy, and maintain governance standards while ensuring duplicate and conflicting tags are identified before creation. #### **Asset Description Updates (update\_asset\_descriptions)** Updates business, technical, domain, tag, and detailed descriptions for supported catalog and governance assets through natural-language requests. The tool enables users to maintain and enrich metadata documentation across one or multiple assets while adhering to existing permissions and governance policies. When descriptions are inherited from associated Business Glossary terms, the tool provides guidance to update the source glossary term to ensure consistency across related assets. ## Authentication The OvalEdge MCP Server uses API Key-based authentication. Generate credentials from within the OvalEdge application and paste them into the MCP configuration file. All MCP requests will run within the security scope of the account that generated the token. Note: Keep your User Secret private. If it's ever compromised, regenerate it from the OvalEdge application. ## REST APIs Available The OvalEdge MCP Server consumes REST APIs for metadata discovery, lineage analysis, glossary retrieval, documentation search, and governance exploration workflows. ### Authentication API
MethodEndpointDescription
POST /user/token/generate Generates a JWT session token for MCP API authentication.
### Metadata & Discovery APIs
MethodEndpointDescription
GET /v1/mcp/search-catalog Performs a hybrid metadata search across catalog assets.
GET /v1/mcp/object-details Retrieves detailed metadata for a catalog object.
GET /v1/mcp/data-sources Returns accessible data source connections.
GET /v1/mcp/column-profile Retrieves profiling statistics for supported assets.
GET /v1/mcp/glossary-terms Retrieves glossary term metadata and definitions.
GET /v1/mcp/tags Retrieves governance tag details and classifications.
### Lineage & Relationship APIs
MethodEndpointDescription
GET /v1/mcp/entity-relationships Returns PK/FK relationship mappings between entities.
GET /v1/mcp/lineage Retrieves lineage graph information for assets.
### Documentation Intelligence API
MethodEndpointDescription
GET /v1/mcp/search-platform-docs Performs semantic search across OvalEdge platform documentation.
## AI and Embedding Integration OvalEdge supports semantic search and vector-based retrieval workflows through integrations with leading AI and embedding providers, including OpenAI, Google Gemini, and Amazon Bedrock. The platform allows organizations to configure and select embedding models based on their specific requirements and retrieval use cases. ## Architecture Diagram ### OvalEdge System Architecture (High-Level View)
### MCP Interaction Flow (Minimal Setup) ![](/files/MOnBJ2oDkPDgr50qD8Jt) ## Open Source MCP Client The OvalEdge MCP server implementation is available as open source on GitHub, where users can find setup instructions, configuration details, and usage examples.\ *** Copyright © 2026, OvalEdge LLC, Peachtree Corners, GA, USA. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.ovaledge.com/release8.1/mcp-server/model-context-protocol.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.