> For the complete documentation index, see [llms.txt](https://agents.glueckkanja.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://agents.glueckkanja.com/troubleshooting.md).

# Troubleshooting

List of common Issues

{% hint style="info" %}
Need help or have questions regarding the agent? E-Mail us at <support.agents@glueckkanja.com>
{% endhint %}

### Agent Won't Run

**"Insufficient privileges" or "Access denied"**

The admin account needs the required Entra ID roles. Check the specific agent's Requirements page and verify all roles are assigned. Most agents need multiple roles (e.g., Compliance Administrator + Security Reader + Reports Reader).

**"Agent not found" or "Cannot load agent"**

Make sure the agent is properly deployed to your Security Copilot environment. Redeploy the agent manifest if necessary.

### Missing or Incomplete Data

**"No data available" or "Insufficient data for analysis"**

Most agents need 30-90 days of historical data to provide meaningful insights. If you've recently deployed a service (Purview, Intune, PIM), wait for data to accumulate before running analysis.

**Agent returns partial results**

Check that all required data sources are configured and collecting data. For example:

* Purview agents need active policies and classification data
* Intune agents need enrolled devices and deployed policies
* PIM agents need role activations occurring

### Permission Problems

**"Cannot access \[service] data"**

The admin who deployed the agent needs access to the specific service. For cross-service agents, verify permissions for all services being analyzed (e.g., Purview + Defender for Compliance Assistant).

**Role assignments not working**

Permissions can take 5-15 minutes to propagate after assignment. Wait a bit and try again. If issues persist, sign out and back in to refresh tokens.

### Authentication Issues

**"Authentication failed"**

Because Entra Agent IDs aren't available yet, agents use the deploying admin's identity. Make sure:

* The admin account has all required roles
* The account hasn't been disabled or locked
* MFA/Conditional Access isn't blocking authentication

### High SCU Consumption

**Agent uses more SCUs than expected**

SCU estimates are based on typical environments. Larger tenants, longer time ranges, or more complex analyses will use more SCUs. To reduce consumption:

* Use shorter time ranges when possible
* For Classification Optimizer, use "quick" mode instead of "deep"
* Limit scope to specific areas (e.g., analyze only DLP instead of all Purview policies)

### Data Quality Issues

**Results don't match what I see in the portal**

* Reports may be delayed 24-48 hours
* The agent and portal might use different data sources or calculation methods
* Check that you're comparing the same time ranges

**Recommendations seem incorrect**

Agents make recommendations based on available data and best practices. They may not have full context about your specific business requirements. Review recommendations carefully and adjust based on your needs.

### Still Stuck?

If none of these help:

1. Check the specific agent's Requirements page for additional notes
2. Verify your Security Copilot subscription is active
3. Ensure the services being analyzed (Intune, Purview, Defender) are properly licensed and configured
4. Review the agent's changelog for known issues in your version


---

# 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://agents.glueckkanja.com/troubleshooting.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.
