# Vectara MCP for AI Agents AI Agent Connect

> Vectara MCP connects your AI agent to your private data corpora for enterprise-grade semantic search and grounded RAG. It lets your agent query indexed documents, manage data sets, and provide cited answers based on your actual internal knowledge. This connection turns your private files into a reliable source of truth for your agent's responses.

## Overview
- **Category:** ai-frontier
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_P55Xdk6UU5XMN3cdVPORquufduaGCfpEWSemsNQg/ai-agent-connect
- **Tags:** rag, semantic-search, enterprise-search, grounded-generation, nlp, data-indexing

## Description

Vectara connects your AI agent to your private data corpora for enterprise-grade semantic search and grounded RAG. Instead of feeding your agent massive chunks of text manually, you can let it pull exactly what it needs through semantic search. It understands the intent behind a question rather than just looking for matching keywords. You can check what's in your data, prune old files that are cluttering up the results, or jump straight into a conversation where the agent provides answers backed by citations from your own files. This is a huge step up for anyone trying to move past basic chat and into actual grounded generation. You can hunt for a specific line in a technical manual or find a needle in a haystack of logs, and this connection ensures your agent stays accurate. You can find this and thousands of other connectors in the Vinkius catalog to get your agent hooked up in minutes. It turns your private corpus into a reliable source of truth for your AI, ensuring that every answer it gives is rooted in your actual data. It also helps data engineers keep their indices clean by removing stale context arrays without having to navigate complex dashboards.

## Tools

### list_corpora
View all searchable datasets in your Vectara account. It's the quickest way to see what's indexed.

### perform_semantic_search
Run a query across one or more corpora. This returns relevant documents based on meaning rather than keywords.

### list_chat_sessions
See a history of previous RAG chat sessions. This helps track past interactions.

### list_corpus_documents
See every document inside a specific corpus. Use this to audit your current data.

### execute_rag_chat
Get a summarized response with citations from a specific corpus. It's perfect for grounded conversations.

### delete_corpus_document
Permanently remove a document from a corpus. This helps keep your search results clean.

### get_corpus_details
Pull metadata and configuration for a specific corpus. Use this to check your data setup.

## Prompt Examples

**Prompt:** 
```
Show me all the corpora I have indexed right now.
```

**Response:** 
```
We retrieved 3 existing corpora within your active Customer ID cluster: 
- 'HR Manuals' (Key: `cor-29x`)
- 'DevOps Runbooks' (Key: `cor-81a`)
- 'Helpcenter FAQs' (Key: `cor-03f`)
Would you like me to pull the total document count indexing actively inside 'DevOps Runbooks'?
```

**Prompt:** 
```
Search the DevOps corpus for anything about kubernetes rollbacks.
```

**Response:** 
```
I've searched corpus `cor-81a`. Here are the top 3 semantic matches:

1. From 'infra-deploy.pdf' (Score 0.94): "To safely rollback a deployed k8s pod replica containing faulty image configs, use `kubectl rollout undo deployment/api-server`."
2. From 'incident-231.md' (Score 0.88): "During yesterday's P1, we reversed the pods deployment to stabilization tag v2.1.0 using standard rollback patterns..."
3. From 'onboarding.txt' (Score 0.76): "New engineers must verify staging before any manual pod cluster rollbacks occur in production environments..."
```

**Prompt:** 
```
What's the history of my RAG chats from last week?
```

**Response:** 
```
I've retrieved historical chat sessions. You have 2 sessions retained matching your request: `chat-89bf` (Initiated: Mon 14th) covering Corpus HR, and `chat-21zx` (Initiated: Wed 16th) heavily focused on developer queries. Do you want to see the sequence of turns for the DevOps troubleshooting session?
```

## Capabilities

### Query private documents with semantic search
Your agent finds relevant documents based on meaning rather than exact keyword matches.

### Get cited answers from your own data
The agent provides summarized responses with direct citations to your source files.

### List and manage your data corpora
You can see all your searchable datasets and their configurations in one place.

### Audit and delete old documents
Remove obsolete files from your index to keep your search results clean and accurate.

### Review previous chat sessions
Access a history of previous RAG interactions to track how the agent handles queries.

## Use Cases

### Debugging RAG responses
A developer uses execute_rag_chat to see how the agent handles a specific query and verify citations without writing a test script.

### Cleaning stale data indices
A data engineer uses delete_corpus_document to remove a stale database schema that's causing the agent to hallucinate.

### Querying internal manuals
A product lead asks questions about the latest manual using perform_semantic_search to get instant answers on internal specs.

### Auditing corpus contents
A technical writer uses list_corpus_documents to find every mention of a specific feature across thousands of embedded files.

## Benefits

- Get cited answers from your own data using execute_rag_chat to ensure your agent stays grounded in facts.
- Keep your search results clean by using delete_corpus_document to prune old or irrelevant files.
- Audit your data environment quickly with list_corpora and list_corpus_documents to see exactly what's indexed.
- Skip the keyword matching headache by using perform_semantic_search to find documents based on actual meaning.
- Track your interactions over time with list_chat_sessions to see how your agent is handling specific queries.

## How It Works

The bottom line is your agent gets immediate access to your private knowledge without you having to copy and paste anything.

1. Subscribe to the Connector and grab your Vectara API keys.
2. Plug those keys into your AI client's configuration.
3. Start asking questions against your indexed data.

## Frequently Asked Questions

**How does Vectara MCP help with RAG?**
Vectara MCP provides a direct connection for your AI agent to query your private data. It enables grounded responses where the agent only answers based on the documents you've indexed.

**Can I use Vectara MCP to search my private files?**
Yes, that is the primary use case. It allows your agent to perform semantic searches across your private PDFs, text files, and other documents stored in your Vectara environment.

**How do I connect Vectara to my AI agent?**
You connect it by adding the Vectara MCP to your supported client like Claude or Cursor. You'll just need your Vectara API Key and Customer ID to get started.

**Can Vectara MCP delete old files from my index?**
Yes, it includes tools to manage your data. You can ask your agent to delete specific documents from a corpus to ensure your search results stay clean and relevant.

**Does Vectara MCP support semantic search?**
Yes, it uses semantic search to find information based on the meaning of your query rather than just looking for exact word matches.

**How do I see my chat history with Vectara?**
The Connector includes a tool to list previous RAG chat sessions, allowing you to review past interactions and queries performed by your agent.

**Can I query my internal documents directly using just conversational chat?**
Yes. If your data is indexed in a Vectara corpus, simply ask your agent: `search the 'employee-handbook' corpus for remote work policies`. The agent uses the `queryTool` to pass your question to Vectara's semantic engine, effortlessly bringing back precisely matching paragraph citations instantly.

**How do I remove outdated context files destroying the accuracy of my RAG model?**
You don't need to rebuild APIs or use cURL. Tell your AI: `delete document ID 'doc-992a' from my Sales corpus`. It automatically formats the mutation and wipes the poisoned embedding from Vectara's nodes permanently, restoring high accuracy.

**Will the RAG Chat tool provide accurate source citations?**
Yes. When you instruct the agent to run `execute_rag_chat`, Vectara processes the query against its internal LLM and index, returning a synthesized natural language answer appended solidly with exact document citations, proving the AI isn't hallucinating facts.