AI Assistant

The Patchly AI assistant lets you ask natural language questions about your security data instead of manually filtering and searching.

How to Use It

1. Click the AI Assistant button in the dashboard
2. Type a question in plain English
3. The assistant queries your data and responds with an answer, often including a table of results

Example Questions

Vulnerability Questions

• “Show me the most exploitable vulnerabilities”
• “Which CVEs are in the CISA Known Exploited Vulnerabilities catalog?”
• “How many critical vulnerabilities were found this week?”
• “What are the top 10 vulnerabilities by EPSS score?”

Device Questions

• “Which servers have the most critical vulnerabilities?”
• “Show critical devices with unpatched CVEs”
• “How many devices are running outdated Windows versions?”

User Questions

• “Which admin users have vulnerable devices?”
• “Show users with the highest criticality scores”

Trend Questions

• “How has our vulnerability count changed over the last 30 days?”
• “Are there any new critical findings since last week?”
• “What changed between yesterday and today?”

How It Works

The AI assistant:

1. Understands your question and translates it into a database query
2. Executes the query against your Patchly data
3. Analyzes the results and provides a natural language summary
4. Shows the underlying data in a table so you can verify

The assistant only has access to your organization’s data and operates within the same permission boundaries as your dashboard.

Tips

• Be specific: “Show critical vulnerabilities on servers” works better than “show problems”
• Ask follow-ups: The assistant remembers context within a conversation
• Request exports: Ask “export these results as CSV” to download data
• Ask about trends: The assistant can compare data across time periods