Skills on a scan

When creating a scan you can attach any number of agent skills. The selected skills are installed into each executor agent for that scan and injected into the prompt as a labelled section.

How to attach

In the scan form's skills section, check the skills you want. There's no limit - pick every skill relevant to the target.

What happens at run time

The chosen skills' Markdown instructions are prepended to the workflow prompt, so the agent applies them on top of the workflow's own steps. Skills complement the workflow; they don't change the required output schema.

Note

Skills are optional. A scan runs fine with none - they're a way to layer reusable, target-appropriate guidance onto a workflow without editing the workflow itself.

Good candidates: language- or framework-specific review guidance, house rules for what counts as a real finding, or checklists for a particular bug class.

Next: post-scripts on a scan.