Weaviate MCP. Semantic Search and Data Governance for AI.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Weaviate connects your AI client directly to a vector database, allowing you to search data by meaning, not keywords. This MCP lets your agent perform deep semantic searches across massive collections of text and objects.
It's built for developers needing to build production-grade applications that require understanding context—whether you’re finding similar documents or auditing cluster health.
What your AI agents can do
Get class schema
Gets the field definitions for one specific data collection in your database.
Get cluster nodes
Checks the operational status and resource usage of all nodes running your Weaviate cluster.
Get full schema
Retrieves a complete map of every class definition across the entire database instance.
Find content that relates semantically to a query vector, even if the original text doesn't contain the exact words used.
Retrieve the complete or partial schema definitions for your entire database instance, allowing you to understand what data fields exist.
List and inspect objects within specific classes, retrieving full property values and metadata.
Retrieve all metadata and internal configurations for a single data object using its unique UUID.
Check the operational status, resource usage (CPU/RAM), and node health of your entire Weaviate cluster.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Weaviate: 7 Available Tools
These tools let you programmatically manage the schema structure, monitor cluster performance, and run highly specific semantic data lookups against your vector collections.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Weaviate on Vinkius019d7620get class schema
Gets the field definitions for one specific data collection in your database.
019d7620get cluster nodes
Checks the operational status and resource usage of all nodes running your Weaviate cluster.
019d7620get full schema
Retrieves a complete map of every class definition across the entire database instance.
019d7620get instance metadata
Pulls high-level configuration and version details about your Weaviate environment.
019d7620get object details
Looks up all the metadata and properties for a single data object using its unique identifier (UUID).
019d7620list objects
Lists records within a specific collection, supporting basic pagination to view multiple entries.
019d7620search near vector
Performs a vector search that finds the closest matching data points based on contextual similarity.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Weaviate, then connect any of our 4,900+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,900+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Weaviate. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 7 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
The Pain of Database Discovery
Today, figuring out what data exists is a slog. You're stuck in a database console, clicking through dozens of tabs and wrestling with nested JSON structures just to understand the available collections or find one specific field definition. It feels like you need an institutional knowledge guide just to start querying.
With this MCP, your agent handles that legwork for you. Instead of manual clicks, you simply ask: 'What are all the classes I have?' The system responds instantly with a map of available data types and structures, letting you move straight to using the information.
Understanding Data Structure with get_full_schema
Before running any complex query, you typically have to guess what fields are available or check documentation that might be out of date. This guesswork slows down development and introduces risk.
By calling get_full_schema, your agent provides a definitive manifest. You immediately know the full data structure—every collection, every property, and its definition—removing all ambiguity from day one.
What you can do with this MCP connector
Forget manually writing complex JSON queries in a database console just to find relevant information. This MCP hands your AI client the ability to treat your vector collection like an extension of its own memory. You can ask questions about massive data sets, and the system finds results based on context and meaning—not just keyword matching.
Your agent gains deep insight into structured metadata; it can list every class in your schema or pull specific object details using a UUID. If you're building an application that needs to connect multiple sources of truth (like pulling user data from a CRM, then checking billing status, and finally sending a message), the whole process runs through Vinkius.
This means even if you need to combine this database connection with other MCPs—say, connecting it to an accounting tool—the entire workflow is secure and visible in one place.
This gives your AI agent full visibility into what every step of the data retrieval process is doing. You don't have to worry about managing credentials or tracking API calls; Vinkius handles all infrastructure and security patches, letting you focus on asking better questions.
019d7620-f2dc-7299-87a2-a2c2f227ec79 
How Weaviate MCP Works
- 1 Subscribe to this MCP and provide your specific Weaviate Base URL and API Key.
- 2 Connect the MCP to any compatible client, like Cursor or Claude. Your agent can now access the database tools through natural conversation.
- 3 Ask your agent a question—for example, 'What are the available collections?' or 'Find articles about quantum computing.' The agent executes the necessary calls and returns structured data.
The bottom line is you use chat prompts to manage complex vector database operations without ever writing code or running manual API commands.
Who Is Weaviate MCP For?
The data engineer who gets bogged down in schema reviews. The research scientist needing quick answers from massive document caches. AI developers building context-aware applications that can't rely on simple SQL lookups.
Auditing database schemas, monitoring cluster health via get_cluster_nodes, and listing objects to validate data ingestion.
Testing semantic search queries using search_near_vector and getting object details to ground the AI's responses in real data.
Quickly surfacing relevant documents from massive, varied vector collections through chat prompts.
What Changes When You Connect
- You can validate your entire system structure by running get_full_schema, which instantly provides a map of every class in the database without manual queries.
- Need to know if an object was correctly indexed? Use get_object_details to inspect all metadata and configurations for any given UUID. This is deep debugging power.
- Instead of keyword searches, use search_near_vector to find information based on context and meaning. It’s essential for building truly intelligent apps.
- Monitor system stability using get_cluster_nodes; you'll know immediately if a node is overloaded or failing before it causes an outage.
- When chaining this MCP with others, your agent can pull data from the database and pass that context to another service, all while Vinkius tracks usage via AI Analytics.
Real-World Use Cases
Finding related articles for a user profile.
A researcher needs to find documents similar to one they just read. They ask their agent to run search_near_vector using the article's vector, which surfaces five highly relevant documents without needing manual query vectors.
Debugging a data pipeline failure.
A developer finds that an object is missing required fields. They use get_object_details by UUID to inspect the full metadata and figure out exactly what properties are wrong, saving hours of guesswork.
Checking database connectivity for a deployment.
An SRE needs to verify if all nodes are running properly. They prompt their agent to run get_cluster_nodes, which immediately reports the CPU and RAM usage across the entire cluster.
Listing available data types in an unknown project.
A new team member joins and needs to know what collections exist. They ask their agent to run get_class_schema or get_full_schema, getting a full list of classes without needing documentation.
The Tradeoffs
Trying to find data with keywords only
Asking the agent to search for 'fast cars' and only getting articles that mention those exact three words, missing related concepts.
→ Use search_near_vector. This function finds items similar in meaning (semantic proximity) rather than just matching specific keywords.
Ignoring operational status checks
Assuming the database is healthy because the last query worked, but failing to notice that one node is spiking on CPU and causing slow reads.
→ Run get_cluster_nodes first. This tells you the actual health status of every node, preventing performance issues before they happen.
Treating metadata as a simple lookup
Trying to find all users by name only, which fails because the database stores complex nested objects and requires specific calls.
→ Use list_objects or get_object_details. These tools allow you to browse collections and inspect full property values for accurate data retrieval.
When It Fits, When It Doesn't
Use this MCP if your primary need is contextual understanding of complex, interconnected data (vector search) or deep system governance (schema/cluster monitoring). If you're building an AI application that must answer 'Why?' or 'What does this mean?', this is the tool. Don't use it if all you need is a simple key-value lookup; for those cases, stick to standard SQL tools. Also, don't rely on get_object_details for general data retrieval; use list_objects first to narrow down which specific UUIDs you need before diving into deep inspection.
Common Questions About Weaviate MCP
How do I find similar documents using search_near_vector? +
You provide a class name and the query vector as requested. The system performs nearest neighbor searches, returning documents that are contextually most relevant to your input.
Can get_object_details tell me what data fields exist? +
It shows the actual metadata for one specific object (by UUID). If you need a list of all possible fields, use get_class_schema or get_full_schema.
What is the difference between listing and searching? +
list_objects shows every record in a class (great for browsing), but search_near_vector finds records based on meaning when you provide a context vector, which is much more powerful.
What does get_cluster_nodes tell me? +
It reports the operational health of your entire cluster, showing CPU and RAM usage for every node. It’s key for monitoring performance and stability.
When should I use `get_class_schema` instead of looking at general metadata? +
It provides the specific property definitions for a single class. Use this when your agent needs to know exactly what fields (like 'title' or 'author') exist in one particular collection before it builds a query.
What information does `get_instance_metadata` provide about my Weaviate environment? +
This tool retrieves high-level operational details. You get the server version number, which modules are enabled, and core configuration settings. This helps confirm that your agent is connected to the expected environment.
How does `list_objects` help me audit data in a specific collection? +
It lists individual data objects within a class, supporting pagination via limit. This lets you sample or browse existing records quickly without needing to formulate a complex vector search query.
What is the value of running `get_full_schema`? +
This tool retrieves the complete schema for every single collection in your Weaviate instance. Use it when you need a comprehensive, global overview before deciding which data types to query.
Multi-server workflows that include Weaviate MCP
Find Codebase Duplications Using MCP Servers
Your codebase has 4 different implementations of date formatting, 3 versions of the retry logic, and 2 competing validation libraries , but nobody knows because grep only finds exact matches and these duplicates are semantic
Improve RAG Search Quality Using MCP Servers
Your RAG retrieves 10 documents but the answer is in #7 , Cohere reranking moves it to #1 and accuracy jumps from 68% to 94% without changing a single embedding
MCP Servers for Self-Updating Research Bases
You spend 3 hours reading 40 articles to write one research brief , an AI agent with Firecrawl reads all 40 in 90 seconds, stores them semantically in Weaviate, and writes the brief in Notion with every source linked and every claim verified
Search Your Entire Codebase Using MCP Servers
Code indexed, patterns detected, architecture documented, onboarding guides generated , build a living knowledge base from your codebase
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.