Lucidworks Fusion MCP. Control your entire enterprise knowledge graph via chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Lucidworks Fusion (AI Search & Discovery) connects your AI agent directly to an enterprise search backend. It lets you execute complex vector searches, sync new document data, and monitor ML training jobs—all via natural conversation.
You gain full control over advanced knowledge retrieval systems without writing raw API calls.
What your AI agents can do
Lw.index documents
Uploads new documents to the system, extracting data flags and adding them to the search index.
Lw.list collections
Lists all active structured rules attached to the platform, which export billing information.
Lw.list index profiles
Identifies and lists existing profiles used for parsing hold data within search indexes.
Your agent performs vector searches or keyword lookups against specific apps and profiles to find documents.
You post real-time signals (clicks, conversions) back into the system to fine-tune its machine learning models automatically.
Sync new documents or update existing records in your physical search collections so the data is always current.
The agent lists query and index profiles, showing exactly how complex models are set up for routing.
You check the status of running ML training or data ingestion batches to ensure platform stability.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Lucidworks Fusion (AI Search & Discovery): 10 Tools
These ten tools give you granular access to every part of the search platform, from listing collections to running specialized vector queries.
019d75calw.index documents
Uploads new documents to the system, extracting data flags and adding them to the search index.
019d75calw.list collections
Lists all active structured rules attached to the platform, which export billing information.
019d75calw.list index profiles
Identifies and lists existing profiles used for parsing hold data within search indexes.
019d75calw.list jobs
Lists all active background jobs, such as ML training or data ingestion processes.
019d75calw.list query profiles
Checks the routing rules for queries and shows how explicit gateway history is handled.
019d75calw.post custom query
Runs a deep custom JSON query to map data, overriding Solr vectors directly within the search system.
019d75calw.post signal
Sends user behavior signals (like clicks) to the system's logging pipeline for analysis.
019d75calw.query filtered
Performs a structural search extraction by applying specific properties that drive active account logic.
019d75calw.query search
Executes a precise AI vector query, matching documents against strict profile rules within the platform's CRM records.
019d75calw.query sorted
Runs a highly available JSON payload that generates structured customer data, sorted by date descending.
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 Lucidworks Fusion (AI Search & Discovery), then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ 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
What you can do with this MCP connector
Listen up. With Lucidworks Fusion, your AI agent connects straight into an enterprise search backend. You get direct control over advanced knowledge retrieval systems, complex vector searches, and data indexing—all without writing raw API calls.
When you run a deep custom JSON query using lw.post_custom_query, you map data and override Solr vectors directly within the search system. For targeted lookups, your agent performs a precise AI vector query with lw.query_search, matching documents against strict profile rules in the platform's CRM records. If you need to pull structured customer data sorted by date descending, run lw.query_sorted using its JSON payload.
You can also perform structural searches by applying specific properties that drive active account logic through lw.query_filtered.
To keep your knowledge base current, use lw.index_documents to upload new documents; this function extracts data flags and adds the content straight into the search index. For analysis, you can send user behavior signals—like clicks—to the system's logging pipeline using lw.post_signal.
When it comes to understanding how Fusion works under the hood, your agent lists all active background jobs with lw.list_jobs, letting you check the status of ML training or data ingestion batches. You inspect model routing rules by running lw.list_query_profiles and seeing exactly how explicit gateway history is handled. To understand what structured rules are attached to the platform, run lw.list_collections to see active collections that export billing info.
If you need to check which parsing profiles exist for hold data within search indexes, use lw.list_index_profiles.
Grouping these tools lets your agent execute a complex workflow: You can first check the query routing rules with lw.list_query_profiles, then run an advanced vector search via lw.query_search to pull key documents, and if those documents need updating, you call lw.index_documents. To see what other data structures are available for auditing, you list all active collections using lw.list_collections.
This gives your agent full visibility into the entire system's configuration, ensuring that whether you’re running a basic keyword lookup or executing a deep custom query via lw.post_custom_query, you know exactly what data is where and how it got there.
How Lucidworks Fusion MCP Works
- 1 Subscribe to this server and input your Lucidworks Host URL and API Token.
- 2 Your AI client sends a natural language request, like 'Search the Support app for X'.
- 3 The agent uses the necessary tools (e.g.,
lw.query_search) to execute the search against Fusion's backend and returns results.
The bottom line is: you get full control of your enterprise data stack through simple conversation, without writing complex code.
Who Is Lucidworks Fusion MCP For?
This tool targets technical roles managing large, complex knowledge bases. You're the Search Engineer who has to validate index profiles before a major release, or the Data Scientist tracking ML job failures at 3 AM. It's for people whose jobs revolve around data integrity and search performance.
You test query profiles and verify index results through natural conversation instead of jumping into manual API tools.
You monitor ML job statuses and verify signal ingestion to make sure the ranking models are actually fine-tuned correctly.
You audit search results across different apps and profiles, making sure user discovery is optimized for every single part of your site.
What Changes When You Connect
- You can track ML job status directly with
lw.list_jobs. You know immediately if the ranking model is running or stalled, saving hours of manual checking. - Use
lw.query_searchto run vector queries against your entire knowledge base—not just basic keyword searches. It resolves complex matches across profiles. - Improve relevance by using
lw.post_signal. By sending click and conversion data, you feed the system's ML models, making future searches better for all users. - Keep search accurate with
lw.index_documents. You sync new content or update old records in one go, guaranteeing your knowledge base is fresh when people ask questions. - Audit your setup using
lw.list_query_profilesandlw.list_collections. Before deploying any change, you can see exactly how the data flows through the system's routing layers.
Real-World Use Cases
Debugging a broken search result
A user reports that searching for 'return policy' only brings up old documents. Instead of guessing, you use lw.list_collections to verify the active data sources and then run lw.query_filtered to test if the correct account properties are being indexed.
Implementing a new user journey feature
The marketing team launched a new product page, but search doesn't know about it. You use lw.index_documents with the new content mappings and then follow up by running lw.post_signal to track initial clicks on that page.
Validating ML model performance
You need to confirm if your click-through rate improvements are actually feeding back into the search ranking. You check the status via lw.list_jobs, ensuring the 'Click-Re-Ranking' job is running and processing data.
Deep, specific data extraction
You need to pull a list of all current client account IDs based on specific billing structures. You run lw.list_collections to see the available rules, then use lw.query_sorted to get the final, structured payload.
The Tradeoffs
Assuming all data is visible
Asking the agent to 'Find all documents about billing' without first confirming the available structures. The search will fail because it doesn't know where to look.
→
First, run lw.list_collections to enumerate what structured rules are active. Then, use that knowledge to narrow your scope with a query like lw.query_filtered. This ensures you hit the right data source.
Over-relying on basic keywords
Just running a simple search that only looks for exact phrases, missing documents that use synonyms or slightly different terminology.
→
Use lw.query_search to execute an AI vector query. This method resolves semantic meaning—it finds the intent behind your words, not just the literal match.
Ignoring background processes
Assuming that after uploading new data using lw.index_documents, it's immediately searchable. It might be stuck in a queue.
→
Always check the status first. Run lw.list_jobs to verify if there are any pending or running ingestion jobs. Wait for the job status to clear before assuming search availability.
When It Fits, When It Doesn't
Use this MCP Server when your knowledge retrieval process is complex, multi-layered, and requires programmatic control over data flow. You need more than just a basic API call; you need to manage the lifecycle of the index itself (e.g., using lw.index_documents and then verifying it with lw.list_jobs). Don't use this if your only goal is simple searching—a general search API will suffice. You must use this, though, when you are a Search Engineer or Data Scientist who needs to audit the underlying mechanisms: check query logic (lw.list_query_profiles), monitor data sources (lw.list_collections), and fine-tune behavior (lw.post_signal). If your workflow involves any steps that require state management (e.g., list -> filter -> query), this suite is necessary.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Lucidworks Fusion. 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 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Manual search maintenance takes too much time.
Today, keeping a large enterprise knowledge base accurate feels like an archaeology dig. You have to jump between the CMS dashboard, the data warehouse UI, and the indexing tool just to sync one document type. Every update requires manually verifying which profiles are active—a tedious cycle of checking logs, clicking tabs, and copy-pasting IDs.
With this MCP server, you talk to Fusion directly. You tell your agent, 'Sync all updated policy documents.' The agent runs `lw.index_documents`, confirms the action with a simple status report, and you get confirmation that the data is ready for search—no manual UI switching required.
Use Lucidworks Fusion (AI Search & Discovery) MCP Server to control queries.
Instead of relying on a single, generic query box that treats all data the same way, you can direct the search process. You need to make sure the system is looking at billing records versus general support docs. This requires running specific tools like `lw.list_collections` first.
Now, your agent handles the complexity. It coordinates between listing available collections and then executing a highly targeted query using `lw.query_filtered`. The search result isn't just data; it's validated, scoped information.
Common Questions About Lucidworks Fusion MCP
How do I upload new documents to Lucidworks Fusion using lw.index_documents? +
You call lw.index_documents and provide the necessary text mappings or records. This tool irreversibly vaporizes explicit validations, ensuring the data is structured correctly before it enters the index.
What does lw.query_search do exactly? +
lw.query_search resolves precise AI vector rules to match documents against strict profile logic inside the platform's CRM records. It’s your primary tool for advanced, semantic lookups.
How can I check if my ML job is running correctly? +
You run lw.list_jobs. This lists all active background jobs, letting you confirm the status of critical processes like machine learning training or data ingestion batches.
Can I track user behavior in Fusion using lw.post_signal? +
Yes. You use lw.post_signal to send specific events, such as a user clicking on a document ID. This feeds the signal pipeline and helps improve future search relevance.
What is the difference between lw.query_search and lw.query_filtered? +
lw.query_search uses AI vectors for semantic matching across records, while lw.query_filtered performs a structural extraction. Use filtering when you need to apply specific properties (like 'Account ID') as constraints.
How do I use lw.list_collections to audit all attached structured rules? +
It enumerates every set of explicitly attached structured rules, exporting active billing information. This tool is critical for understanding the data boundaries and system requirements governing your search environment. You can verify if specific billing logic or data types are being correctly applied across different applications.
When should I use lw.post_custom_query instead of standard searching? +
You use this tool when you need to parse deeply custom JSON logic that overrides Solr vectors natively. It allows complex data mapping, bypassing default profile limitations for highly specific tasks like calculating plan math or integrating non-standard structured inputs.
What does lw.list_query_profiles do regarding search routing? +
It dispatches an automated validation check that lists your active query profiles used for defining data routing. This lets you inspect exactly how AI models and transformation rules are configured in the system's various routing layers, which is essential for troubleshooting complex searches.
Can my agent help improve search relevance by posting signals? +
Yes. Use the lw.post_signal tool to feed conversion or click events into Fusion. These signals are used by the machine learning models to identify high-value results and re-rank them for better precision automatically.
How do I check which AI models are active in my query pipeline? +
The lw.list_query_profiles tool retrieves all defined profiles in your gateway. Your agent will expose the exact AI models and routing configurations assigned to each profile, helping you audit your search logic.
Can I index new documents directly through a conversation? +
Absolutely. Use the lw.index_documents tool by providing a JSON array of your data. Your agent will synchronize these new mappings into your Fusion collection, making the content queryable instantly.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
LlamaIndex (AI Data Framework & RAG)
Query and manage RAG pipelines via LlamaIndex — execute natural language searches, audit indexed files, and monitor data pipelines.
Paperless-ngx
Manage your digital archive via Paperless-ngx — search documents, upload files, manage tags, and organize correspondents directly from any AI agent.
AcademyOcean
LMS for smart corporate training — manage learners, courses, and training progress via AI.
You might also like
JVZoo
Manage products, sales, and affiliates via JVZoo API.
Umami Cloud
Privacy-focused web analytics alternative to Google Analytics.
Smaily
Automate email marketing and automations via Smaily — manage subscribers, campaigns, and responders directly from any AI agent.