> For the complete documentation index, see [llms.txt](https://docs.logilica.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.logilica.com/metrics-and-reports/ai-insights.md).

# AI-Powered Insights

Logilica includes an AI assistant that analyses your dashboard data and explains it in plain English. Instead of interpreting charts yourself, open the AI panel to get a summary you can paste into a Slack message, a retro document, or a stakeholder update.

## How to Use It

1. Open any dashboard in Logilica
2. Click the **sparkle icon** (floating button in the bottom-right corner of the screen) — this opens the AI chat panel
3. The AI immediately generates a **summary of your current dashboard** — this appears automatically
4. Below the summary, you'll see **suggested prompts** — click any of them to get a deeper analysis on that topic
5. The AI analyses the data behind the charts currently shown on the dashboard and streams a response in real time

Each prompt can be used once per session. If you change the dashboard filters (time range, team, project), the AI automatically re-generates a fresh summary based on the updated data.

## Available Prompts

When the AI panel opens, you'll see these prompts after the initial summary:

| Prompt                                                                                                     | What It Does                                                                                                        |
| ---------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------- |
| **"Can you provide me with more details about this dashboard?"**                                           | Gives a deeper breakdown of the data across all chart tiles — trends, comparisons, and notable values               |
| **"Which areas within this data set present the most significant opportunities for improvement?"**         | Identifies metrics that are underperforming or declining, and highlights where the team could focus effort          |
| **"What immediate actions can we take to enhance our team's efficiency, as indicated by this dashboard?"** | Suggests specific, actionable next steps based on what the data shows — bottlenecks to address, processes to adjust |

Each prompt is tailored behind the scenes to produce a detailed, structured response. The label you see is a simplified version — the actual instruction sent to the AI is more specific.

## What the AI Sees

The AI has access to the **data behind the charts on your current dashboard**, including:

* The metric values and trends shown in each chart tile
* The filters you've applied (time range, team, project)
* Table data if the dashboard contains table tiles

The AI analyses the same data you see on screen. It does not access raw code, individual commits, or information beyond what the dashboard displays.

## What the AI Is Good At

* **Summarising** key data points across multiple charts into a single narrative
* **Explaining** what the numbers mean in context — translating metrics into plain English
* **Identifying** areas that need attention — declining trends, outliers, or bottlenecks
* **Suggesting actions** based on the data — concrete next steps rather than abstract observations

## What the AI Doesn't Know

The AI analyses the data shown on your current dashboard — it doesn't have context about your team's circumstances. A cycle time spike might be caused by a holiday week, a team reorg, or a major incident — the AI sees the numbers but doesn't know about these events. You bring the context; the AI brings the data summary.

## Tips

* **Set your dashboard filters before opening the panel** — the AI analyses whatever data is currently shown. Narrow the time range and select the right team first, otherwise the summary may be too broad to be useful
* **Use all three prompts for a complete picture** — the initial summary gives you the overview, "more details" breaks it down, "opportunities" shows where to focus, and "immediate actions" tells you what to do next
* **Copy and paste the output** — AI responses are formatted for sharing. Paste them directly into Slack, a retro doc, or a stakeholder email
* **Change filters to get a different perspective** — switch the team or time range filter and the AI automatically re-generates its summary. Use this to quickly compare different teams or time periods
* **AI summaries also appear in weekly email reports** — if AI insights are enabled for your organisation, the weekly email report includes AI-generated summaries alongside the charts. See [Reports](/metrics-and-reports/reports.md) for details


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.logilica.com/metrics-and-reports/ai-insights.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.