• INDEX
  • FILES
  • GAP
  • HUMAN

How to verify document ingestion status with your AI

Connect LlamaIndex to your AI client once. From there, you can audit your data pipeline and confirm your indexes match your source files.

Ask AI about this page

Short answer

How can I check if my AI has ingested all my documents?

You can use your AI to compare the list of files in your storage against the active semantic indexes. This confirms that every source file is accounted for and correctly indexed. You'll see exactly what's indexed and what's missing.

Audit outcomes

Where your data status lands.

Running these checks gives you three specific ways to validate your data.

INDEX

Verify semantic indexes

The AI pulls a list of all active indexes. You'll know exactly which vector stores are ready for retrieval.

FILES

Check raw source files

The AI scans your storage to list every file present. This helps you spot files that haven't been processed yet.

GAP

Identify missing data

By comparing files to indexes, your AI flags discrepancies. You'll see which documents failed to ingest.

HUMAN

Manual cleanup required

If the AI finds a mismatch, you'll need to manually re-run your ingestion script or fix the source file. The AI cannot fix broken data pipelines.

The workflow

What your AI does when the ingestion check arrives.

Your AI acts as an auditor for your LlamaIndex data structures.

  1. Inventory indexes

    The AI identifies all existing semantic structures currently managed by LlamaIndex.

    list_indexes

  2. Scan source files

    The AI looks at your raw data storage to see what files are actually sitting there.

    list_files

  3. Compare datasets

    The AI cross-references the file list with the index list to find gaps.

    list_indexeslist_files

  4. Report discrepancies

    The AI presents a clear summary of which files are indexed and which are ignored.

    list_files

Try it

Copy these to start.

Use these prompts to begin your audit.

Starting points

These are basic requests. Swap out the folder names or index names to match your specific setup.

LlamaIndex Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_0oOSMctRD8JekU5V9qPIM95pb3koS0z8x0sprJvK/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — LlamaIndex capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "llamaindex-ai-data-framework-rag-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_0oOSMctRD8JekU5V9qPIM95pb3koS0z8x0sprJvK/mcp"
    }
  }
}
Copy into chat04

  • List all my current LlamaIndex indexes so I can see what's active.

  • Show me all the files in my 'documents/finance' folder.

  • Compare my active indexes against the files in my 'data/raw' directory.

  • Are there any files in my source folder that aren't part of an index yet?

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect LlamaIndex once, then ask.

One connection gives your AI access to your data framework. Your credentials stay encrypted, and you can start auditing immediately.

Connect LlamaIndex to your AI

FAQ

How this task behaves

  • 01

    Can the AI fix an index that failed to build?

    No. The AI can only list what exists and what is missing. You must handle the actual rebuilding of indexes in your LlamaIndex environment.

  • 02

    Can the AI delete files from my storage?

    No. This MCP only provides tools to list indexes and files. It cannot modify, delete, or move your data.

  • 03

    How does the AI know if a file is correctly indexed?

    The AI compares the file list from your storage against the index list. If a file exists in storage but isn't in an index, it flags it as unindexed.

  • 04

    Can I see the content of the files through the AI?

    No. These tools only allow the AI to list the files and indexes. It does not have permission to read the actual text inside the files.

  • 05

    Does this work with any vector database?

    This works with any data managed through the LlamaIndex framework that these specific tools can access.

  • More questions about LlamaIndex? The Connector page answers them. See everything the LlamaIndex Connector can do

Connect LlamaIndex to Claude, Cursor, ChatGPT & more