# Weaviate MCP for AI Agents AI Agent Connect > Weaviate MCP lets you manage your vector database and perform semantic searches directly through your AI client. It connects your Weaviate instance to your agent so you can query data by meaning rather than just keywords. It handles everything from checking cluster health to pulling specific object details or inspecting your full schema. ## Overview - **Category:** loved-by-devs - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_ewBpKaEj3GWOCB4HdPc1Tag8b8QBrbGJDpCpFWfW/ai-agent-connect - **Tags:** semantic-search, vector-embeddings, schema-management, nearest-neighbor, ai-native, data-indexing ## Description Managing a vector database usually means jumping between a database console and your code, constantly checking if your nodes are healthy or if your schema is actually what you think it is. This Connector changes that by putting the entire Weaviate instance under your agent's control. Instead of writing manual JSON queries to find a specific object or checking the status of your cluster nodes, you just ask. You can browse your collections, inspect the metadata of specific records, and run similarity searches using raw vectors in a single conversation. It makes your AI agent act like a dedicated database administrator. Since Vinkius hosts this Connector, you can get it running in your favorite client in minutes. It's about moving from "how do I query this?" to "find me something like this" without leaving your chat window. ## Tools ### get_cluster_nodes Retrieve operational status and resource usage for your cluster nodes. It provides a quick health check for your infrastructure. ### get_object_details Pull all metadata and properties for a specific object using its UUID. This is great for deep auditing of individual records. ### get_full_schema Fetch the complete schema for every collection in your Weaviate instance. Use this to get a bird's-eye view of your entire data structure. ### list_objects Browse and list data objects within a specific class. It includes basic pagination to help you explore your collections. ### search_near_vector Perform a nearest neighbor similarity search using a provided class name and vector array. This is how you find contextually relevant data. ### get_class_schema Get the specific schema definition for a single collection class. This helps you confirm what properties are available for your queries. ### get_instance_metadata View high-level metadata about your current Weaviate instance. Use this to check your active modules and version details. ## Prompt Examples **Prompt:** ``` What's the current status of my Weaviate cluster? ``` **Response:** ``` ### Cluster Health Report | Node | Status | CPU Usage | RAM Usage | | :--- | :--- | :--- | :--- | | Node-1 | **HEALTHY** | 42% | 6.5GB | | Node-2 | **HEALTHY** | 38% | 6.2GB | | Node-3 | **HEALTHY** | 40% | 6.4GB | All nodes are operating within normal parameters. No issues detected. ``` **Prompt:** ``` Show me the schema for the 'Products' class. ``` **Response:** ``` ### Products Class Schema Here are the properties defined for the **Products** collection: * **id**: UUID (Primary Key) * **name**: String * **description**: Text * **price**: Float * **category**: String * **vector**: [Float] (Vector Embedding) ``` **Prompt:** ``` Find items similar to this vector: [0.12, -0.05, 0.88] ``` **Response:** ``` I found 3 similar items in the **Articles** collection: 1. **Vector Search for Beginners** (Score: 0.94) 2. **Understanding Embeddings** (Score: 0.89) 3. **Intro to Vector DBs** (Score: 0.82) Would you like to see the full content for the top match? ``` ## Capabilities ### Run similarity searches Find relevant data based on vector embeddings and context. ### Inspect database schemas View your full collection structure or specific class definitions. ### Monitor cluster health Check the status and resource usage of your Weaviate nodes. ### Browse data objects List items within a class to see their properties and values. ### Pull specific records Retrieve deep metadata for a single object using its UUID. ### Check instance details View your current Weaviate version and active modules. ## Use Cases ### Verifying successful data ingestion A developer needs to see if a specific record was ingested correctly. They ask the agent to pull the details of a UUID using get_object_details. ### Checking infrastructure health An SRE notices a lag in search results. They ask the agent to check the health of all nodes using get_cluster_nodes. ### Exploring new collections A researcher wants to see what's in a new collection. They ask the agent to list the objects in that class using list_objects. ### Confirming schema definitions An engineer needs to confirm the schema of a production class. They ask the agent to fetch the definition using get_class_schema. ## Benefits - Skip manual JSON queries by using search_near_vector to find similar content through natural language. - Monitor your infrastructure in real time with get_cluster_nodes to see node health and CPU usage. - Verify your data structure instantly with get_full_schema to ensure your classes are mapped correctly. - Audit specific records quickly using get_object_details to inspect metadata without opening a database console. - Simplify schema management by using get_class_schema to pull specific definitions for individual collections. - Get a bird's-eye view of your setup with get_instance_metadata to check versions and active modules. ## How It Works The bottom line is you get a conversational interface for your vector database without the manual overhead of complex queries. 1. Add your Weaviate Base URL and API Key to the Connector configuration. 2. Connect your AI client to the Weaviate MCP via Vinkius. 3. Ask your agent to perform searches or check cluster stats in plain English. ## Frequently Asked Questions **Can I use Weaviate MCP to find similar items in my database?** Yes. You can ask your agent to find content similar to a specific concept or a provided vector. It uses the similarity search tool to pull the most relevant results for you. **How do I check my cluster health using the Weaviate MCP?** Simply ask your agent for a health report. It will automatically check your cluster nodes and summarize the CPU, RAM, and status of your infrastructure. **Can the Weaviate MCP show me my full database schema?** Yes, it can pull the entire schema for all your collections. This is helpful for getting a clear overview of your data structure in one go. **Is the Weaviate MCP good for auditing specific records?** It's excellent for auditing. You can provide a UUID to your agent and it will pull every piece of metadata and property associated with that specific object. **Can I use the Weaviate MCP to see my current Weaviate version?** Yes. You can ask for instance metadata, and the agent will report back the version, active modules, and high-level configuration details. **Does the Weaviate MCP support searching by vector embeddings?** Yes, that is a core feature. You can provide a vector array to your agent, and it will perform a nearest neighbor search to find the best matches. **Can I perform a vector search using float arrays through the agent?** Yes. The `search_near_vector` tool allows you to perform semantic searches by providing a query vector as a JSON array of floats. Your AI agent will return the most similar objects from your Weaviate collection. **How do I see the data structure of my Weaviate collections?** You can use the `get_full_schema` tool to see all classes and properties defined in your instance, or `get_class_schema` if you want to focus on a specific collection's definition. **Is it possible to monitor cluster health via chat?** Absolutely. Use the `get_cluster_nodes` tool to retrieve operational data for all nodes in your Weaviate cluster, including their current status and resource utilization metrics.