> For the complete documentation index, see [llms.txt](https://docs.uptiq.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/troubleshoot-evals.md).

# Troubleshoot Evals

Fix missing scores, evaluators you can't create, and review tasks that don't reach anyone. Contact support if your problem isn't listed.

Find the cause of a missing score, an evaluator you can't create, or a review task that doesn't reach anyone. If your problem isn't listed, see [Get help](#get-help).

## Missing scores

### An agent's conversations aren't being scored

**Issue**. Conversations for an agent have no scores.

**Cause**. Assigning evaluators doesn't start scoring on its own. **Online eval** starts off, and a low **Sampling** value excludes most conversations.

**Solution**. Check all three settings.

1. Open **Agent List** and confirm evaluators are assigned to the agent.
2. Turn on the agent's **Online eval** toggle.
3. Raise **Sampling** if it's set too low.

See [Assign evaluators](https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/agent-list.md#assign-evaluators).

### An agent shows Unsupported instead of the Online eval toggle

**Issue**. The agent's row shows **Unsupported** where the **Online eval** toggle should be.

**Cause**. That agent's type doesn't support Online Eval.

**Solution**. Pre-assign evaluators with **Edit evaluators** so they apply when the agent type supports Online Eval. See [Agent List](https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/agent-list.md).

### A conversation doesn't appear in Online Eval Runs

**Issue**. A conversation is missing from Online Eval Runs.

**Cause**. Online Eval Runs lists runs that have been scored. The agent has no evaluators, **Online eval** is off, or **Sampling** excluded the conversation.

**Solution**. Check the agent's configuration.

1. Confirm the agent has evaluators assigned.
2. Confirm **Online eval** is on.
3. Check that **Sampling** isn't excluding conversations.

See [Online Eval Runs](https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/online-eval-runs.md).

### A score has no reasoning or evidence

**Issue**. **View details** on a run's row shows no **Evidence** list in **Reasoning**.

**Cause**. The judge returned no reasoning or evidence.

**Solution**. Nothing to fix. See [Judge evidence](https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/online-eval-runs.md#judge-evidence).

## Evaluators and review

### New evaluator isn't available

**Issue**. **+ New evaluator** isn't available, or you see "You don't have access to create or change evaluators."

**Cause**. Your role lacks the **Create & enable evaluators per agent** capability. Developer, Governance Manager, Account Admin, and Organisation Admin have it.

**Solution**. Ask an administrator for one of those roles. See [Roles](https://docs.uptiq.ai/ai-workbench-qore/control-center/admin/roles.md).

### Create evaluator isn't available

**Issue**. **Create evaluator** in the new evaluator dialog isn't available.

**Cause**. The dialog needs a name, a method, and a judge model or jury models. A checklist needs at least one check.

**Solution**. Complete the missing items.

1. Enter an **Evaluator name**.
2. Choose the judge model, or two to five jury models.
3. For a checklist, add at least one check.

See [Create an evaluator](https://docs.uptiq.ai/ai-workbench-qore/task-guides/create-an-evaluator.md).

### Low-confidence results never reach the Review Queue

**Issue**. Results the judge is unsure about don't appear in the Review Queue.

**Cause**. The evaluator's **Review rule** reads **Off**, so none of its results go to review.

**Solution**. Turn routing back on.

1. Open the evaluator's **Actions** menu and select **Edit**.
2. Turn on **Review when the judge is unsure**.
3. Select **Save changes**.

See [Create an evaluator](https://docs.uptiq.ai/ai-workbench-qore/task-guides/create-an-evaluator.md).

### A reviewer sees no tasks under Assigned to me

**Issue**. A reviewer's **Assigned to me** view is empty.

**Cause**. Tasks aren't assigned automatically.

**Solution**. Assign tasks from the Review Queue.

1. Open the [Review Queue](https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/review-queue.md) and leave **Status** on **Needs review** to see unassigned tasks.
2. Select the tasks, and then select **Assign to me**. Alternatively, open a task and choose an assignee.

## Get help

If your problem isn't listed, contact your Uptiq administrator. Include the agent and evaluator names, the time you expected a score or task, and what the screen showed.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.uptiq.ai/ai-workbench-qore/control-center/evals/troubleshoot-evals.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
