Skip to main content
Overview is the first item in the agent sidebar. It needs calls from a connected agent, and the outcome, intent and red flag numbers depend on the call classification and red flag rules you’ve set up.
Each agent has an Overview dashboard for a daily health check, and analysis dashboards for a closer look. See Analysis dashboards.

​
Overview

The Overview dashboard summarizes your agent’s performance over the period you select. Use it to tell whether the agent is broadly healthy with isolated issues, or degraded across the board. Tuner Overview Dashboard

​
Headline metrics

Five metrics sit at the top. Pick the period with the 7 days, 30 days or 90 days buttons. Each card also shows the change against the previous period of the same length. When there are matching calls, Failed Calls and Calls with Red Flags show a View calls link that opens Call Logs filtered to those calls. Success Rate and Failed Calls depend on how you’ve defined call outcomes. If these numbers look wrong, check those definitions before debugging the agent. Simulation calls are not counted in the Overview metrics.
Per-call values like duration and cost aren’t headline metrics. They appear on each call in Call Logs, and you can use them as conditions in red flag rules.

​
Breakdowns

Two charts show how calls are spread across your own definitions:
  • Call Intent Breakdown: the mix of reasons people called.
  • Call Outcome Breakdown: the mix of results.
Read them together. An intent with high volume and a poor outcome share is where to spend your time first. The MCP diagnosis follows the same reasoning.

​
Recent red flags

The Recent Red Flags table lists recent calls that tripped one of your rules, with the FLAG TYPE, CALL ID, CALL INTENT, OUTCOME and TIMESTAMP. Click View to open the call. See Red flags and labels for how to work through the list. Red flags only appear here if you’ve defined rules. An empty list on a busy agent usually means no rules exist yet.

​
Diagnose your agent

A banner at the top of the dashboard reads “Run a full diagnosis to trace red flags, hallucinations and latency back to a root cause.” Click Diagnose Agent to open the Diagnose agent dialog. It shows how to connect Tuner to your AI client and gives you a prompt to paste there, already filled in with your agent and workspace IDs. The same button is on Call Logs. The button is disabled until the agent has calls. See Diagnose your agent for the full workflow.

​
Analysis dashboards

Three more dashboards hold detailed charts for the selected agent:

​
Next steps

Call Logs

Search, filter and open the calls behind the numbers.

Diagnose your agent

Use the MCP to find why something is wrong across every affected call.