• COMPLIANCE
  • INSIGHTS
  • RECOVERY
  • HUMAN

How to audit conversation history with your AI

Connect your DynamoDB tables to your AI client to run deep scans across your entire chat history. Once connected, your agent can pull any record to find exactly what you're looking for.

Ask AI about this page

Short answer

How can I audit my chat logs using AI?

Your AI uses the scan_table tool to read every entry in your DynamoDB table. It then parses those logs to find specific keywords, user behaviors, or compliance issues. You get a filtered list of relevant records.

Audit outcomes

Where your findings land.

The AI processes raw data into actionable categories.

COMPLIANCE

Flagged policy violations

The AI identifies messages that break your specific rules. It pulls these records into a list for you to review.

INSIGHTS

User behavior patterns

The agent looks for recurring themes or sentiment shifts across thousands of logs. You get a summary of how users are actually interacting.

RECOVERY

Missing record detection

The AI compares expected logs against what is actually in the table. It highlights gaps where data might be missing.

HUMAN

Manual review queue

The AI marks high-priority items that require a person to step in. You decide which flagged logs need a human response.

The workflow

What your AI does when the scan starts.

The AI handles the heavy lifting of reading raw data so you don't have to.

  1. Data retrieval

    The agent performs a full sweep of your DynamoDB table to pull all existing chat records.

    scan_table

  2. Pattern matching

    The AI compares the retrieved logs against your specific audit criteria or keywords.

    scan_table

  3. Anomaly detection

    Your agent identifies outliers, such as sudden spikes in certain topics or unusual user activity.

    scan_table

  4. Report generation

    The AI organizes the findings into a structured format that highlights the most important issues.

    scan_table

Try it

Copy these to start.

Use these prompts once you have connected your table.

Starting points

These are basic requests. Replace the table names and keywords with your actual DynamoDB details.

Amazon DynamoDB Table Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_n32klzizAPnc4YO7K3qG23La7StLiOV91J7uGn2a/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 — Amazon DynamoDB Table capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "amazon-dynamodb-table-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_n32klzizAPnc4YO7K3qG23La7StLiOV91J7uGn2a/mcp"
    }
  }
}
Copy into chat04

  • Scan the ChatLogs table and find any messages containing 'refund' from the last 24 hours.

  • Review the UserInteractions table to identify any users who have sent more than five messages in a minute.

  • Look through the SupportHistory table for any logs where the sentiment is highly negative.

  • Perform a scan of the SessionData table to find all entries where the 'status' field is set to 'error'.

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

Start here

Connect Amazon DynamoDB Table once, then ask.

Just provide your credentials through the secure connection interface. Once it's live, your AI can immediately start scanning your tables.

Connect Amazon DynamoDB Table to your AI

FAQ

How the audit process behaves

  • 01

    Can the AI delete or change my chat logs?

    No. The AI only has the ability to read data via the scan tool. It cannot modify, delete, or overwrite any records in your DynamoDB table.

  • 02

    Does the AI store my data?

    No. Your data stays in your DynamoDB table. The AI only accesses it during the execution of a tool call to answer your specific request.

  • 03

    How does the AI handle very large tables?

    The AI uses the scan tool to iterate through the table. It processes the records to find matches based on your prompt.

  • 04

    Can I use this to update user profiles?

    No. This specific task is for auditing and reading history. There are no tools provided to write or update data in your table.

  • 05

    Will the AI automatically run these scans every day?

    No. The AI only performs a scan when you explicitly ask it to do so in your chat interface.

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

Connect Amazon DynamoDB Table to Claude, Cursor, ChatGPT & more