Skip to main content
The dashboard shows which calls are failing. A diagnosis explains why. The Tuner MCP server reads your workspace and can write back to it, so the same assistant that finds the root cause can add the evals, intents and data fields that measure it, and suggest the prompt change.
Diagnosis needs the Tuner MCP server connected to your AI client, and calls from the last 30 days on the agent.

​
Run a diagnosis

Paste this into your chat to run a full diagnosis:
Replace agent_id, workspace_id, and agent_name with your own values. To skip typing them, click Diagnose Agent on the Overview page to copy the prompt with your values filled in. You can also ask first: “List the agents in workspace 1420”. See API keys and IDs for where the workspace ID is.
Diagnose Agent button in the Tuner dashboard, prefilled with a prompt to run a full diagnosis

​
What the diagnosis looks at

Following the tuner_analyze_agent prompt, the assistant works through your data in this order:
1

Reads your agent's configuration

Pulls the system prompt, workflow, and description, plus everything you’ve defined: outcomes, intents, data-extraction fields, evals, and red flags. Everything else is measured against this.
2

Builds the health picture

Reviews success rate, red-flag rate, call volume, and trends over the last 30 days, then breaks calls down across your outcome labels and intent categories, to tell whether the agent is broadly healthy with isolated issues or degraded across the board.
3

Finds the priority signals

Looks at recently red-flagged calls and which intents and outcomes they cluster on. A red flag concentrated on one intent with bad outcomes becomes the lead to chase.
4

Isolates the problem segments

Filters the call logs down to the underperforming intent + outcome combinations and checks whether the problem is chronic or started at a specific agent version. A change that lines up with a version points to a regression.
5

Digs into individual calls

Opens the worst calls in detail: which evals are failing, whether STT, LLM, TTS, or end-to-end latency is elevated, and whether the same red flags fire consistently.
6

Cross-references against your prompt

Compares what the failing calls show against what your prompt and workflow are actually designed to do. The gap between the two is the root cause.

​
What you get back

The assistant returns a structured diagnosis:
  • Agent health: a one-line summary with the headline number that supports it.
  • Issues found: each with the Signal (the data, with numbers), the Pattern across failing calls, the Root cause hypothesis and confidence, a specific Fix (what to change and where), and how to Verify it worked.
  • What’s working: so you know what not to touch.
  • Data gaps: configuration that, if added, would sharpen future diagnoses.

​
From diagnosis to fix

The same assistant can apply the fix in the same conversation. The MCP server’s write tools create, update and delete call outcomes, intents, data-extraction fields, evals, capabilities, tools, red-flag and label rules, and alerts, and can update the agent’s name, description, language and call direction. See What the MCP server can do. After it explains the root cause, tell it to act. It can:
  • Draft the prompt and workflow change for you to make in your agent.
  • Create or refine evals so the behavior is scored on every future call.
  • Add call outcomes and intents so the right calls get categorized.
  • Add data-extraction fields to capture the values you need to measure the problem.
  • Add red flags and alerts so you’re warned when it happens again.
Then diagnose again to confirm the fix. Each pass adds the measurement the next diagnosis needs.

​
Examples

​
A spike in unresolved calls on one intent

Suppose your Billing Dispute intent is ending Unresolved far more often than the rest of your calls. Ask:
“Diagnose my agent: why are so many Billing Dispute calls ending unresolved, and how do I fix it?”
The assistant pulls the outcome breakdown and sees Billing Dispute sitting at a 48% Unresolved rate vs. 12% agent-wide. It opens the worst calls and finds the same pattern: the agent can’t move forward because it never collects the account number before trying to act on the dispute. It reports the signal, the pattern, the root cause, and a concrete fix. Then you close the loop in the same chat:
“Draft the account-verification step for my prompt, create an eval that checks the agent collected and verified the account number, and add a data field to capture it.”
The assistant drafts the prompt change for you to apply, and through the MCP server’s write tools creates the eval (“Did the agent verify the account number before discussing the dispute?”) and the data-extraction field. Every future Billing Dispute call is now measured on that behavior, so your next diagnosis can confirm whether the Unresolved rate is dropping.

​
Finding problems you didn’t know to look for

You can only write evals for problems you already know about. Diagnosis can surface the others. Ask:
“Are there calls that don’t match any of my configured intents? Cluster them and tell me what they’re about.”
The assistant scans your call logs for calls that landed with no matching intent, reads the transcripts, and groups them into themes. For example, 62 calls in the last 30 days asking about international shipping, an intent you never defined. Because there’s no flow for it, those calls fall through to a generic response and abandon at a high rate. The assistant explains the theme, shows you example transcripts, and recommends:
  • Add an International Shipping intent so these calls are tracked.
  • Add a prompt or workflow branch that handles the request.
  • Add an eval to check the agent gives accurate shipping information.
You can have it create the intent and eval in the same conversation.

​
Keep asking

After the diagnosis, you can keep asking about your calls in plain language:
“Show me the transcripts of the 3 worst calls behind that issue.”
“What’s different about the calls that fail versus the ones that succeed on this intent?”
“Break down latency for those calls. Which component is slow: STT, the LLM, or TTS?”
“Did this start at a specific agent version, or has it always been like this?”
“Draft the exact prompt change you’re recommending, then create the eval to track it.”
“Which eval is failing the most this week, and on what kind of calls?”
Diagnosis needs real calls to analyze. If your agent hasn’t received any calls in the last 30 days, the assistant will ask you to send calls through it first. With fewer than 10 calls it will still run, but it flags that the findings may not yet be statistically reliable.

​
Next steps

Set up the MCP server

Connect the Tuner MCP server to your AI client and configure your agent.

Red flags and labels

The red flags the diagnosis prioritizes when triaging issues.