> ## Documentation Index
> Fetch the complete documentation index at: https://docs.idemeum.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Run dashboard
> Overview of AI agent runs.
## What is a run?
A run is one attempt to fix a problem. Assist matches a skill, builds a plan, checks it against the guardrails, runs the steps with the employee's approval, verifies the result and records it as a ticket. Questions and follow-ups are answered in the conversation without starting a run.
Runs start from a typed request, a quick action or a [proactive trigger](/assist-proactive-actions). Each run records:
* **Goal:** the request, in the employee's words.
* **Skill:** the playbook that handled it, such as `disk-cleanup` or `triage`.
* **Plan and steps:** each step's tool, inputs, result and outcome.
* **Approvals:** every preview shown and every change approved or declined.
* **Feedback:** whether the employee said the problem was fixed.
* **Summary:** the result in plain language.
## Run dashboard
In the idemeum admin portal, go to **Runs**. The overview covers the last 30 days.
| Metric | What it shows |
| ---------------------- | -------------------------------------------------------------------------------------------------- |
| Total | All runs in the period. |
| Resolved | Runs Assist fixed end to end, and their share of the total. |
| Escalated | Runs handed to IT because the employee said the problem wasn't fixed. |
| Failed | Runs that couldn't complete: an error, a stopped run, or no step able to run. |
| MTTR | Average time Assist spent working on a run, not counting time waiting for the employee's approval. |
| Successful run rate | Runs where every step succeeded. |
| CSAT | Positive ratings, and their share of all ratings. |
| Weekly resolution rate | Share of runs resolved each week, over the last 90 days. |
Successful run rate and Resolved measure different things. Every step can succeed while the problem persists, and a run can be resolved even if the employee declined a step. A gap between the two means fixes are running but not always helping.
After you turn on or edit a skill, watch the weekly resolution rate; the effect shows within a week or two.
## How a run's result is decided
Every run ends as **Resolved**, **Escalated** or **Failed**. What the employee says outranks what the steps reported:
1. **Failed:** the steps didn't complete.
2. **Escalated:** the steps ran, but the employee says the problem isn't fully fixed.
3. **Resolved:** the steps ran, and the employee confirmed the fix or didn't answer.
```mermaid theme={null}
%%{init: {
"theme": "base",
"themeVariables": {
"fontFamily": "Inter, ui-sans-serif, system-ui, sans-serif",
"fontSize": "14px",
"primaryColor": "#ffffff",
"primaryBorderColor": "#e4e4e7",
"primaryTextColor": "#18181b",
"lineColor": "#a1a1aa",
"edgeLabelBackground": "#ffffff"
},
"flowchart": { "useMaxWidth": false, "nodeSpacing": 40, "rankSpacing": 42 }
}}%%
flowchart TD
Q1{"Did the steps
complete?"} -- no --> F("Failed")
Q1 -- yes --> Q2{"Did the employee say
the problem persists?"}
Q2 -- yes --> E("Escalated")
Q2 -- "no, or no answer" --> R("Resolved")
classDef card fill:#ffffff,stroke:#e4e4e7,stroke-width:1px,color:#18181b;
classDef primary fill:#18181b,stroke:#18181b,color:#fafafa;
classDef destructive fill:#fef2f2,stroke:#fecaca,stroke-width:1px,color:#b91c1c;
classDef muted fill:#f4f4f5,stroke:#e4e4e7,stroke-width:1px,color:#3f3f46;
class Q1,Q2 card;
class R primary;
class E muted;
class F destructive;
```