Security Audit¶
The meterian:security-audit skill audits your project's open-source dependencies
for known vulnerabilities and offers to fix them.
What it does¶
Mode A — Full project audit¶
When you ask "audit my dependencies", "scan this project", or simply open a manifest file, the skill:
- Finds all manifest files in the workspace
- Resolves exact installed versions, from lock files when available (including transitive dependencies), or from manifest version ranges otherwise
- Sends all dependencies to the Meterian CLI in one batch call
- Presents results as a table with severity, CVE ID, and safe versions
## Meterian Security Audit Report
| Package | Version | Severity | ID | Safe Versions |
|---------|---------|----------|----------------|---------------|
| lodash | 4.17.15 | HIGH | CVE-2021-23337 | 4.17.21 |
1 vulnerability found across 42 packages (41 clean).
If no vulnerabilities are found: ✅ No vulnerabilities detected across N packages.
Mode B — Ad-hoc query¶
Ask "Is express 4.17.1 safe?" or "What CVEs affect log4j 2.14.1?" — the skill looks up that specific package and reports inline.
Mode C — Remediation¶
After an audit, the skill offers to fix vulnerable packages:
- Patch-level upgrades are applied automatically
- Minor or major upgrades are shown first and require your confirmation
After applying fixes, the skill re-runs the audit to confirm all packages are clean.
Auto-trigger manifest files¶
The skill activates automatically when you open or modify any of these files:
package.json · package-lock.json · yarn.lock · pnpm-lock.yaml ·
requirements.txt · Pipfile · pyproject.toml · poetry.lock · uv.lock ·
pom.xml · build.gradle · build.gradle.kts · Cargo.toml · Cargo.lock ·
composer.json · composer.lock · Gemfile · Gemfile.lock ·
go.mod · go.sum · *.csproj · conanfile.txt · conanfile.py ·
pubspec.yaml · pubspec.lock · project.clj · deps.edn ·
Package.swift · Package.resolved
Relationship to Reachability Analysis¶
After finding vulnerabilities, the skill offers to run a Reachability Analysis to determine which of them are actually exploitable in your codebase.