• REPORT
  • ALERT
  • HUMAN
  • LOG

How to audit administrative actions with your AI

Connect CodeRabbit to your agent to pull direct logs from your system. Once connected, you can ask your agent to scan for specific changes or unauthorized permission shifts.

Ask AI about this page

Short answer

How do I audit admin actions with my AI?

Your agent uses the get_audit_logs tool to pull raw, tamper-resistant logs directly from your environment. You can then ask it to identify specific permission updates or suspicious organizational changes. You'll get a clear view of what happened.

Audit outcomes.

Where your findings land.

The AI processes the logs and organizes the results into these categories.

REPORT

Clean audit summaries

The AI pulls the logs and writes a clear summary of what happened. You get a readable list of changes without digging through raw text.

ALERT

Flagged permission shifts

The agent identifies when a user's access level changes. It highlights these specific events so you don't miss them.

HUMAN

Manual review required

If the AI finds a conflict or an unknown user, it stops and asks you to verify the action. You decide if the change was legitimate.

LOG

Raw data exports

You can ask the agent to format the logs into a specific structure. This makes it easy to move the data into other tools.

The workflow

What your AI does when the logs arrive.

The agent handles the heavy lifting of data retrieval and pattern matching.

  1. Fetch logs

    The agent reaches into your system to grab the latest administrative event data.

    get_audit_logs

  2. Filter changes

    It scans the retrieved data for specific keywords like 'permission' or 'admin'.

    get_audit_logs

  3. Identify anomalies

    The agent looks for patterns that don't match your usual organizational activity.

    get_audit_logs

  4. Summarize findings

    It condenses the raw log entries into a concise report for your review.

    get_audit_logs

Try it

Copy these to start.

Use these prompts to begin your audit.

Starting points

Swap out the bracketed text for your specific repo name or user ID.

CodeRabbit Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_PMcRlSWUfPE40MnPxowwrCoT2UjR21cc7zb3FshU/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 — CodeRabbit capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "coderabbit-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_PMcRlSWUfPE40MnPxowwrCoT2UjR21cc7zb3FshU/mcp"
    }
  }
}
Copy into chat04

  • Get the audit logs for the production repo and list all recent permission changes.

  • Check the logs for any admin actions performed by user 'jdoe' in the last 24 hours.

  • Scan the recent audit logs for any unauthorized changes to organizational settings.

  • Summarize all administrative actions taken on the main branch since Monday.

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

Start here

Connect CodeRabbit once, then ask.

Just click the link to connect. Your credentials stay encrypted, and you can start auditing immediately through your preferred AI client.

Connect CodeRabbit to your AI

FAQ

How this task behaves.

  • 01

    Can the AI change permissions itself?

    No. The AI can only read the logs via get_audit_logs. It cannot modify any settings or permissions in your system.

  • 02

    Does the AI delete any audit history?

    No. The AI has read-only access to the logs and cannot delete or alter any existing data.

  • 03

    How recent is the data the AI sees?

    The AI sees whatever is available in the logs provided by the get_audit_logs tool at the moment you run the command.

  • 04

    Can I use this to fix errors?

    No. This tool is for auditing and visibility. You will need to manually log in to your system to correct any issues the AI finds.

  • 05

    What happens if the logs are too large?

    The AI will process the logs it can retrieve. If the log volume exceeds the tool's limits, you may need to ask for specific time ranges.

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

Connect CodeRabbit to Claude, Cursor, ChatGPT & more