> For the complete documentation index, see [llms.txt](https://imbrace.gitbook.io/imbrace-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://imbrace.gitbook.io/imbrace-documentation/governance/conversation-logs.md).
# Conversation Logs
Conversation Logs (AI Tracing) lets you review how AI agents processed individual conversations. Use it for troubleshooting, quality checks, and understanding why an AI agent responded in a particular way.
Logs can be browsed in two ways:
* **By Session:** Review the trace for a single chat session.
* **By Agent:** Review sessions handled by a specific AI Agent within a date range.
### **Viewing Logs by Session**
1. Click the **Conversation Logs (AI Tracing)** tab.
2. Click **By Session**.
3. Click the **Chat Session** drop-down and select a session.
4. Once a trace is loaded, the page displays summary cards, aggregate performance metrics and a list of AI Agent responses.
5. Click **REFRESH** to load the trace again for the selected session.
> **Note:** The **REFRESH** button is only available after a chat session is selected.
### **Viewing Logs by Agent**
1. Click the **Conversation Logs (AI Tracing)** tab.
2. Click **By Agent**.
3. Click the **Use Case (Agent)** drop-down and select an AI Agent. To find an AI Agent quickly, enter its name in the **Search** field within the drop-down. Once results are loaded, the page displays summary cards, latency metrics and tool performance for the selected AI Agent.
4. To search for a specific response of AI agent in a certain time period, add the dates in **From** & **To** field and search. Click **APPLY** to load the logs.
> **Tip:** Use **By Agent** when investigating an issue reported for a specific AI Agent, then narrow the date range to the period when the issue occurred.
