Skip to main content

📝 Execution Logs

Every time a tool is run, MCP for WP records a detailed log of the event. These logs are an invaluable resource for debugging issues, auditing usage, and monitoring the performance of your AI tools.

🔍 What Information is Logged?

For each execution, the following details are captured and stored:

  • Tool ID and Name: Identifies which tool was executed.
  • Timestamp: The exact date and time the tool was run.
  • Status: The final outcome of the execution (Success or Error).
  • Request Payload: The full input data (JSON) that was sent to the AI provider.
  • Response Payload: The full raw output (JSON) that was received from the AI provider.
  • User Information: The WordPress user who initiated the execution (if applicable).

🖥️ How to View the Logs

You can easily access the execution logs from your WordPress admin dashboard.

  1. Navigate to Logs: Go to MCP for WP > Logs.
  2. Review the List: The screen displays the 100 most recent executions in a clear, table-based format.

Logs Screen Screenshot

The log viewer includes columns for the time, the tool that was run, the status, and the request/response payloads.

🛠️ Common Use Cases for Logs

Debugging Failed Executions

This is the most common use for the logs. If a tool fails, the Response Payload will contain the exact error message returned by the AI provider's API. This could be anything from an invalid API key to a problem with the prompt itself. This detailed feedback is crucial for diagnosing and fixing the issue.

Auditing Tool Usage

The logs provide a complete audit trail of all tool activity. This can be useful for understanding which tools are being used most frequently, who is using them, and when they are being run.

Monitoring API Performance

By reviewing the timestamps and status of your logs, you can get a sense of how quickly the AI provider is responding and whether there are any intermittent performance issues.

✅ Best Practices

  • Check Logs First: If you encounter an issue with a tool, the execution logs should always be the first place you look.
  • Regularly Review: It's a good practice to periodically review the logs to check for any unexpected errors or patterns.
  • Understand the Payloads: Familiarize yourself with the structure of the request and response payloads for your tools. This will make it much easier to debug issues when they arise.