> For the complete documentation index, see [llms.txt](https://docs.nexthink.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nexthink.com/user-guide/spark/working-with-spark/using-recommendations-to-enhance-spark-knowledge-and-actions.md).
# Using Recommendations to enhance Spark knowledge and actions
Nexthink Spark resolves employee issues based on the knowledge and actions available to it. When these are missing, Spark cannot consistently resolve issues.
Spark **Recommendations** automates the identification of these gaps. It analyzes thousands of employee conversations with Spark each week and surfaces prioritized recommendations to your IT team, starting with the highest-impact issues.
Each recommendation identifies the root cause behind a failing topic, such as missing or outdated knowledge or required Agent actions that aren’t enabled. It then explains what to fix, why it matters, and which conversations drove the finding. You can explore insights using the same Nexthink Assist capabilities available in Workspace, all within the Recommendations view.
## Accessing Spark Recommendations
To access Spark Recommendations, select the Spark module from the main menu and click **Recommendations** in the navigation panel. This opens a page listing all recommendations on the left panel, and chat interface with details on the right.
## Reviewing recommendations
Recommendations on the left side panel are grouped by assignee (**Assigned to me**, **Unassigned**, **Assigned to others**).
* Review the recommendation titles with a clear, actionable description of what needs to be fixed.
* Preview recommendation details below the title: status, how long ago it was created, and the number of escalations caused by this gap.
* Hover over a recommendation to view the full description and assignee.
### Filtering and sorting capabilities
Use the search box to find a recommendation based on keywords or phrases, such as the application name or type of issue.
Filter the list of recommendations by the following properties:
* **Assignee** (Assigned to me, Unassigned, Assigned to others),
* **Status** (Open, Done, Dismissed)
* **Type** (Action missing, Knowledge missing).
Click **Reset** to clear all filters and see all recommendations
Sort by **Priority** (highest number of escalations first) or **Last updated** (most recent changes first).
## Acting on a recommendation
New recommendations that you have not yet reviewed and that have not yet been assigned are marked with a **blue dot**.
To review and act on a recommendation, click the selected recommendation to view detailed information, including what employees are asking, why Spark is not resolving it, and the recommended next step.
### Digging deeper with Nexthink Assist
If you need more context before taking action, use the chat area at the bottom of the details panel to ask questions about the issue. The chat is powered by **Nexthink Assist** and helps you explore the recommendation’s context and potential solutions in more detail.
Refer to [Using Workspace](/user-guide/search-and-workspace/using-workspace.md) and [Understanding Nexthink Assist](/user-guide/search-and-workspace/understanding-nexthink-assist.md) for more information.
{% hint style="info" %}
Starting a chat automatically assigns the recommendation to you, but other Nexthink accounts can preview its details.
{% endhint %}
### Assigning Spark recommendation
You can assign a selected recommendation to yourself or someone else:
**Manual assignment**
* To assign to yourself: Click the assignee field at the top and select your name.
* To assign to someone else: Click the assignee field, use the search box to find a Nexthink account, and click to assign
**Automatic assignment through chat**
If the recommendation is unassigned, starting a chat with Nexthink Assist automatically assigns it to you.
{% hint style="info" %}
If the recommendation is already assigned to someone else, you cannot chat but can preview the assignee’s chat conversation to see their findings.
{% endhint %}
### Changing recommendation status
Update the recommendation status to track your progress and keep the team informed of what’s being worked on. You can change a recommendation to **Done** or **Dismissed**.
**Mark as Done**
Once you’ve addressed the gap, click **Mark done**:
* If it is a **knowledge gap**: You have created or updated the KB article and confirmed Spark is connected to it
* If it is an **action gap**: You have enabled the recommended action for Spark
Refer to [Managing Spark settings and data inputs](/user-guide/spark/setting-up-and-managing-spark/managing-spark-settings.md) and [Managing Spark actions](/user-guide/spark/setting-up-and-managing-spark/managing-agent-actions.md) for more information.
**Dismiss**
Click **Dismiss** to close a recommendation without making improvements. Dismiss only when you have determined the recommendation should not be implemented.
***
RELATED PAGES:
* [Monitoring Spark](/user-guide/spark/working-with-spark/monitoring-spark.md)
* [Setting up and managing Spark](/user-guide/spark/setting-up-and-managing-spark.md)
* [Using Workspace](/user-guide/search-and-workspace/using-workspace.md)
---
# 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.nexthink.com/user-guide/spark/working-with-spark/using-recommendations-to-enhance-spark-knowledge-and-actions.md?ask=&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.
