Nuclei Template Validation Pipeline
How the automated Helix → Nuclei pipeline validates CVE researcher-generated detection templates against real targets before they reach the detection fleet.
Overview
CVE researcher-generated Nuclei templates are automatically validated against real targets when a pull request is opened in the Nuclei templates repository. Only templates that successfully detect the vulnerability on a live target are permitted to merge. This eliminates the manual validation step that previously existed between template authorship and fleet deployment.
How the pipeline works
1. Pull request triggers validation
When a researcher opens a PR against the Nuclei templates repository, the pipeline starts automatically. No manual action is required to initiate validation.
2. Target retrieval from Helix
The pipeline reads the Linear ticket associated with the PR and retrieves the affected target hostname from the ticket metadata in Attack Helix. This removes any manual handoff between the research workflow and the validation environment — the target context flows directly from Helix into the pipeline.
3. Template execution against a real target
The Nuclei template is run against the hostname retrieved in the previous step. Validation uses a real, affected target rather than a synthetic or mocked environment, so results reflect actual detection behavior.
4. Merge gate enforcement
Validation passes — the PR is cleared to merge and the template proceeds to the detection fleet.
Validation fails — the PR is blocked from merging automatically. The researcher is expected to review the template, address any issues, and push an updated commit to re-trigger validation.
Researcher workflow
Prerequisites
For the pipeline to retrieve a target hostname, the following must be true before the PR is opened:
A Linear ticket for the CVE exists and is linked to the PR.
The ticket contains a valid target hostname in its Helix-sourced metadata.
The target host is reachable from the validation environment at the time the pipeline runs.
If any of these conditions are not met, the pipeline will be unable to resolve a target and the PR will be blocked until the ticket metadata is corrected.
Related modules
Emergent Threat / CVE Automation — broader CVE detection automation context within Guard.
Attack Surface Management — how target hostnames are tracked and maintained across the platform.