Voice AI Agent Logs helps you review, trace, and debug Voice AI interactions from one place in HighLevel. You can use Agent Logs to inspect individual calls, follow agent handoffs, review timestamps, and identify where a conversation succeeded, slowed down, or failed. This article explains what Agent Logs shows, how to use it, and what to check when you are troubleshooting Voice AI behavior.
TABLE OF CONTENTS
- What is Voice AI - Agent Logs?
- Key Benefits of Voice AI - Agent Logs
- How To Navigate and Use Agent Logs in Voice AI
- Frequently Asked Questions
- Related Articles
What is Voice AI - Agent Logs?
Voice AI Agent Logs gives you detailed visibility into how a Voice AI interaction was handled from start to finish. It helps you review the conversation, inspect execution events, confirm agent transfers, and understand what happened at each step of the call. This makes it easier to troubleshoot unexpected responses, failed calls, and slow actions without relying on guesswork.
Key Benefits of Voice AI - Agent Logs
Conversation visibility: Review the agent conversation, contact responses, and call activity in one place.
Execution tracing: Follow each message and action with exact timestamps to understand how the call progressed.
Transfer debugging: See which agent triggered a transfer, which agent took over, and whether the handoff was successful.
Payload inspection: Toggle between raw and parsed data to review what was sent and returned during a step.
Latency review: Identify slower steps and understand where time was spent during execution.
Model auditing: View which model handled a step in the conversation.
Failure visibility: Surface call failure reasons directly when a call or agent step does not complete successfully.
How To Navigate and Use Agent Logs in Voice AI
Agent Logs is designed to help you quickly move from a high-level call list to the exact step that needs attention. Following the full review flow below makes it easier to find the right log, inspect the interaction, and isolate transfer issues, payload details, latency, or failure reasons in a single workflow.
Navigate to AI Agents > Agent Logs.
Use the logs table to locate the interaction you want to review. Depending on what is available in your account, you can review details such as:
Agent name
Contact name
AI Product
Channel
Timestamp
Status
Actions
Use the available filters to narrow down results as needed:
Click on the Agent's name to view details or on the three dot Action button and select on View Details option.
Review the Agent conversation view to understand how the interaction unfolded. This view can include:
Call recording playback
The back-and-forth conversation between the contact and the agent
Quick access to execution details tied to a message or event
Use the Execution Timeline to step through the interaction event by event. Each timeline item includes an exact timestamp so you can identify where the conversation changed, slowed down, transferred, or failed.
Select a timeline item to inspect its details. Depending on the step, the detail panel can show:
Model used: to audit behavior or compare execution quality across interactions.
Type
Latency: to identify where time was spent. This is useful when a Voice AI interaction feels delayed or inconsistent.
Timestamp
Input
Output
Metadata
When reviewing Input or Output, use the toggle to switch between Raw and Parsed views:
Raw shows the original payload for the selected step.
Parsed shows a more readable version of that payload when available.
If the interaction includes an agent transfer, review the timeline to confirm the full handoff chain. Agent Logs can help you verify:
Which agent triggered the transfer
Which agent took over the conversation
Whether the transfer was successful
What input and output were recorded for the transfer event
If call recording playback is not available, verify whether call recording is enabled for the Twilio number in LC Phone settings. Playback depends on call recording being enabled for that number.
Frequently Asked Questions
Q: Why is call recording playback missing?
Verify whether call recording is enabled for the Twilio number in LC Phone settings. Playback depends on call recording being enabled for that number.
Q: What can I use Agent Logs for?
Agent Logs helps you review Voice AI interactions, inspect step-by-step execution, debug transfers, review payload data, check latency, and identify failure reasons.
Q: What is the difference between the conversation view and the Execution Timeline?
The conversation view helps you read the interaction as it happened, while the Execution Timeline helps you inspect each event and action behind the interaction in more technical detail.
Q: What does Raw versus Parsed mean in Input and Output?
Raw shows the original payload for the selected step, while Parsed shows a more readable version of that data when available.
Q: Can I see which model handled a step?
Yes. When you select a timeline item, the detail panel can show the model used for that step.
Q: Can I track agent transfers in Agent Logs?
Yes. Agent Logs can show which agent triggered the transfer, which agent took over, and whether the transfer succeeded.
Q: Why is call recording playback missing?
Call recording playback is only available if call recording is enabled for the Twilio number in LC Phone settings.
Q: Can Agent Logs help me troubleshoot slow Voice AI behavior?
Yes. Per-step latency helps you identify slower parts of the interaction so you can see where time is being spent.
Q: What should I check when a call fails?
Open the log detail, review the Execution Timeline in timestamp order, inspect the output for the failed step, and look for the surfaced failure reason.
Related Articles
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article