> For the complete documentation index, see [llms.txt](https://docs.ox.security/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ox.security/vibesec/agent-ai-bom/skills.md).

# Skills

## Skills overview

You can use the Skills page to discover AI agent skills across your organization and assess their security. VibeSec scans skills available to developers and identifies malicious behavior and vulnerabilities.

<figure><img src="https://884876233-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FdK3XMLdV8zRg847RmGmZ%2Fuploads%2Fgit-blob-116143cb8ff36e3971d62545541d5a1cb7dad970%2Fskills_general.png?alt=media" alt=""><figcaption></figcaption></figure>

VibeSec identifies each skill by its name and hash. The hash provides a unique identifier because multiple skills can have the same name.

If a skill is modified, its hash changes. As a result, different versions of a skill with the same name appear as separate entries on the Skills page.

## How skills are discovered

VibeSec discovers skills automatically when a user starts a new agent session.

For each session, VibeSec scans:

* Global skills available to the agent on the user's device.
* Repository-specific skills associated with the repository where the session is started.

Skills in other repositories are scanned when a user starts an agent session in those repositories.

A skill does not need to be actively used during the session to be discovered.

## Skills table

The Skills page lists the skills discovered across your organization and their security assessments.

| Column     | Description                                                                                                                                                                                                                                                                                                                                                                                                                 |
| ---------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Name       | Name of the skill and its unique hash identifier.                                                                                                                                                                                                                                                                                                                                                                           |
| OX Verdict | <p>OX assessment of the skill, as follows:<br><br><strong>Dangerous:</strong> The skill is identified as malicious.<br><br><strong>Borderline dangerous:</strong> Automated analysis identified behavior that may be malicious, but the assessment has not been confirmed.<br><br><strong>Safe:</strong> No malicious behavior or vulnerabilities were identified.<br><br>A skill can be both Malicious and Vulnerable.</p> |
| Category   | Purpose or integration associated with the skill, such as AI / ML, GitHub, or Notion.                                                                                                                                                                                                                                                                                                                                       |
| Findings   | Number of security findings identified for the skill.                                                                                                                                                                                                                                                                                                                                                                       |
| First Seen | When VibeSec first scanned the skill.                                                                                                                                                                                                                                                                                                                                                                                       |
| Last Seen  | When VibeSec most recently scanned the skill. The skill does not need to be actively used to update this value.                                                                                                                                                                                                                                                                                                             |

#### Filters

You can use the following filters to narrow the skills displayed in the table:

| Filter     | Description                                                                                                                      |
| ---------- | -------------------------------------------------------------------------------------------------------------------------------- |
| Category   | Filters by the skill category, such as code development, testing and QA, DevOps and CI/CD, file management, GitHub, or database. |
| Findings   | Filters by whether security findings were identified for the skill.                                                              |
| Name       | Filters by skill name.                                                                                                           |
| OX Verdict | Filters by the security assessment assigned to the skill.                                                                        |
| User       | Filters by users associated with the skill.                                                                                      |

You can also use Columns to select which columns appear in the table.

#### Borderline skills

When VibeSec identifies a skill as potentially malicious, a warning indicates that the assessment is based on automated analysis and has not been confirmed.

To investigate an individual skill, select the skill from the table. For more information, see Skill Details.

### Enabling skill scanning

Skill scanning is controlled from the VibeSec settings.

After skill scanning is enabled, VibeSec scans applicable skills when users start new agent sessions. Scan results appear on the Skills page.

You can also review skill scan activity in the Agent Activity Log by filtering the activity type for skill scanning events.

**To enable skill scanning:**

1. **Go to Settings > VibeSec.**
2. **Enable Skill Audit & Scan.**

## Related Topics:

[AI Inventory](https://docs.ox.security/vibesec/agent-ai-bom.md)

[Skill Details](https://docs.ox.security/vibesec/agent-ai-bom/skills/skill-details.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.ox.security/vibesec/agent-ai-bom/skills.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.
