- INDEX
- SEARCH
- ANSWER
- REVIEW
Short answer
How can I make my AI search my own documents?
You connect VectorShift to your AI client to enable semantic searches across your uploaded files. Your agent finds the right context in your wikis to answer questions accurately. You'll see how the data moves from indexing to retrieval below.
The outcomes.
Where your data lands.
The AI follows these steps to turn your files into usable answers.
INDEX
Data is prepared
The AI organizes your documents into a searchable format. This allows it to find meaning instead of just matching keywords.
SEARCH
Context is found
When you ask a question, the AI scans your indexed knowledge to find the specific passages that matter.
ANSWER
Facts are delivered
The AI uses the retrieved text to give you a direct answer based on your own data.
REVIEW
Human check
You review the AI's response against the source material to ensure the context is right before you use it.
The workflow
What your AI does when the request arrives.
The AI handles the heavy lifting of finding and reading your files.
Organize files
The AI takes your raw documents and turns them into a structured, searchable index.
index_knowledge_baseLocate sources
The AI looks through your existing collections to see what information is available.
list_knowledge_basesExtract answers
The AI performs a semantic search to pull the exact information needed for your prompt.
query_knowledge_baseVerify context
The AI uses the retrieved snippets to construct a response that stays grounded in your data.
query_knowledge_base
Try it
Copy these to start.
Use these prompts to test the connection immediately.
Starting points
These are opening lines. Swap the bracketed text for your actual file names or wiki titles.
VectorShift Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_EmQ0ykbVFnJJJOx8Wrsr9PBOJ4yiohOhm84Pa96G/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — VectorShift capabilities are ready to use.
{
"mcpServers": {
"vectorshift-ai-workflow-rag-automation-mcp": {
"url": "https://edge.vinkius.com/vk_preview_EmQ0ykbVFnJJJOx8Wrsr9PBOJ4yiohOhm84Pa96G/mcp"
}
}
}Search the [Company Wiki] for our remote work policy.
Based on the [Product Manual], how do I reset the device?
Find the specific details about [Project X] in my uploaded docs.
Summarize the key points from the [Q3 Strategy] knowledge base.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect VectorShift once, then ask.
Click the button to link your account. Your credentials stay encrypted, and the full walkthrough lives on the Connector page.
Connect VectorShift to your AIFAQ
How this task behaves.
- 01
Can the AI edit my original documents?
No. The AI can only read and query the information you have indexed.
- 02
How does the AI find information if I don't use exact keywords?
It uses semantic search, which means it understands the meaning behind your question.
- 03
Does the AI delete any of my files?
No. This MCP only provides read and index capabilities; it cannot delete your data.
- 04
Can I see a list of all my indexed bases?
Yes, the AI can list your available knowledge bases to see what it has access to.
- 05
Can the AI create new documents in my wiki?
No. The tools only allow for indexing, listing, and querying existing knowledge.
More questions about VectorShift? The Connector page answers them. See everything the VectorShift Connector can do
Connect VectorShift to Claude, Cursor, ChatGPT & more
