Praetorian uses AI agents to generate Nuclei templates for CVE detection

What happened and what to do

The article from Praetorian, published on 26 February 2026, describes a multi-agent pipeline that starts from CVE research and produces validated Nuclei detection templates. Specialized agents divide the steps of the work, the system uses three model providers, and an actor-critic refinement loop improves the results before validation.

For a company, the pattern matters to any team that needs to respond quickly to new vulnerabilities. A typical implementation would start with automated collection of advisories and research, followed by template generation in an isolated environment, validation against test targets, and human review before publishing to internal scanners. The models can run on the client's infrastructure, with logs for each step, coverage metrics and a dashboard tracking false positives and the time between advisory and detection.

How the consultancy can help

We diagnose the current CVE response workflow and design a pipeline for collecting, generating and validating detection templates, implemented and operated on the client's infrastructure with human review at the critical steps.

Next step

If your security team wants to shorten the time between a vulnerability's disclosure and its detection, send a short description of your case. We will reply with a scoped proposal.