- STRUCTURE
- ACCURACY
- QUERY
- REVIEW
Short answer
How do I build a RAG knowledge base with my AI?
Upload your files and organize them into a knowledge graph through the Writer MCP. This structure lets your agent query your private data to answer questions accurately. You'll see how the data moves from raw files to a searchable graph below.
The results of your connection
Where your data goes
Once you've connected your files, your AI gains these capabilities.
STRUCTURE
Organized knowledge graphs
The AI turns loose documents into a connected graph. This makes it easy to find relationships between different pieces of information.
ACCURACY
Grounded answers
Your agent uses the graph to pull facts from your files. This prevents the AI from making things up when you ask about your data.
QUERY
Direct data access
You can ask questions directly against your uploaded files. The AI scans the graph to find the exact context it needs.
REVIEW
Human oversight
You decide which files are added to the graph. The AI only knows what you explicitly upload and organize.
The workflow
What your AI does when you provide data.
The AI handles the heavy lifting of structuring your information so you can just ask questions.
Initialize the graph
The AI sets up a new container to hold your structured information.
create_graphIngest your files
Your documents are sent to the system to be processed.
upload_fileBuild relationships
The AI maps the content from your files into the existing graph structure.
add_file_to_graphRetrieve answers
When you ask a question, the AI searches the graph for the relevant context.
ask_question
Try it
Copy these to start.
Use these prompts to trigger the workflow immediately.
Starting points
These are starting points. Swap out the bracketed text with your actual file names or graph names.
Writer Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_yrsful81rgtIcMZXt4yQE12SkSrbuSWmhu7UnyRz/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 — Writer capabilities are ready to use.
{
"mcpServers": {
"writer-ai-enterprise-llm-mcp": {
"url": "https://edge.vinkius.com/vk_preview_yrsful81rgtIcMZXt4yQE12SkSrbuSWmhu7UnyRz/mcp"
}
}
}Create a new knowledge graph named 'Project Alpha'.
Upload the document 'Q3_Report.pdf' to my current graph.
Add 'employee_handbook.docx' to the 'Company Policy' graph.
Based on the files in my graph, what is our policy on remote work?
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Writer once, then ask.
One click connects the MCP to your client. Your credentials stay encrypted, and you can find the full technical details on the Connector page.
Connect Writer to your AIFAQ
How this task behaves
- 01
Can the AI delete my original files?
No. The AI only interacts with the files through the upload and graph tools. It cannot delete files from your local machine or cloud storage.
- 02
Does the AI automatically read everything in my drive?
No. The AI only sees the data you explicitly upload or add to a graph using the provided tools.
- 03
Can I use this with Claude?
Yes, as long as you are using an MCP-compatible client like Claude Desktop.
- 04
Will the AI update my files if it finds an error?
No. The AI can only read and add data to the graph. It cannot modify or overwrite your original source documents.
- 05
How does the AI find answers in a large graph?
The AI uses the graph structure to navigate between related pieces of data to find the specific context needed for your question.
More questions about Writer? See everything the Writer Connector can do
Connect Writer to Claude, Cursor, ChatGPT & more
