Agentic Product Lifecycle

mxa-audit-security

Assesses a codebase and its running site for security by driving the standard scanners, and scores only what a scanner covered. It ships no scanner of its own.

  • Runs five categories of check with industry-standard tools: vulnerable dependencies, leaked secrets in the tree and its history, static analysis, response headers and a passive scan of the running site
  • Chooses the static-analysis rules from the languages in the tree, so a React app and a Rails app get different rules from the same skill
  • Scores only the categories a scanner ran, and states the unseen share as the ceiling, so a high number can never mean nobody looked
  • Writes two reports, one for the client and one to work through, and sends a leaked key to rotation rather than deletion
  • Runs at the start of an engagement, before a promotion and on a live estate, and never becomes a gate

What is inside

The checks module drives gitleaks, OSV-Scanner, Semgrep and the site scanners and maps their reports onto the artifact. It scans nothing itself.

Uses
gitleaksOSV-ScannerTrivySemgrepBrakemanMozilla ObservatoryOWASP ZAP
Evals
3

Works with

The outside tools and connectors this skill drives.

  • gitleaks· our default
  • OSV-Scanner or Trivy
  • Semgrep, with Brakeman on Rails
  • Mozilla Observatory
  • OWASP ZAP baseline

The first is what we reach for on our own projects. A project that already uses one of the others keeps it, and the skill works with that instead.

Install it

npx skills add madebymany/madebyagents --skill mxa-audit-security
Read the skill itself