Skip to content
Back to skills

Dependency Scan

ASecurity

Audit dependencies for known CVEs and license risks

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 17, 2026
ai-agentsgogitdatabasesecurity

Security analysis

A100/100

Scanned October 17, 2026

npx -y skills add bromso/metapowers --skill dependency-scan --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Dependency Scan?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Dependency Scan
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/bromso-dependency-scan/badge)](https://www.skillsdirectory.com/skills/bromso-dependency-scan)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
description: Audit dependencies for known CVEs and license risks
---

# Dependency Scan

Audit dependencies for known CVEs and license risks for "$ARGUMENTS". Enumerate all direct and transitive dependencies, identify vulnerabilities and licensing concerns, and produce actionable remediation guidance.

## Prerequisites

Read `.metapowers/security/$ARGUMENTS/00-govern.md`. If this file does not exist, tell the user:

> Phase 0 (Govern) has not been completed for "$ARGUMENTS". Run a Govern skill first (e.g., `/security:security-policy $ARGUMENTS`), or use `--skip-checks` to bypass.

If `--skip-checks` is present in $ARGUMENTS, skip this check and log to `.metapowers/security/$ARGUMENTS/skip-log.md`.

## Process

1. **Enumerate all dependencies:**
   - Identify package manifests (package.json, requirements.txt, go.mod, Cargo.toml, pom.xml, etc.)
   - List all direct dependencies with pinned versions
   - Resolve transitive dependency tree to identify hidden risks
   - Note any vendored or manually included libraries

2. **Check against CVE databases:**
   - Cross-reference each dependency against NVD (National Vulnerability Database)
   - Check GitHub Advisory Database for known security advisories
   - Review language-specific advisory sources (npm audit, pip-audit, cargo-audit, etc.)
   - Score each vulnerability by CVSS (Critical >= 9.0, High 7.0-8.9, Medium 4.0-6.9, Low < 4.0)

3. **Identify license risks:**
   - Catalog the license of each dependency (MIT, Apache 2.0, GPL, LGPL, AGPL, BSL, etc.)
   - Flag copyleft licenses (GPL, AGPL) in commercial or proprietary projects
   - Identify dependencies with no declared license or custom license terms
   - Flag unmaintained packages (no commits > 2 years, archived repositories, no response to security issues)

4. **Produce remediation recommendations:**
   - For each vulnerable dependency recommend: **patch** (minor update), **upgrade** (major version), or **replace** (alternative library)
   - Assess breaking change risk for each recommended action
   - Prioritize by CVSS score and exploitability (known exploits in the wild rank higher)
   - Identify dependencies that can be removed entirely

5. **Create dependency policy:**
   - Define auto-merge rules for patch updates (automated via Dependabot, Renovate, etc.)
   - Define manual review requirements for major version updates
   - Set maximum age threshold for dependencies without security updates
   - Establish approved license list and blocked license list

6. **Write the artifact** to `.metapowers/security/$ARGUMENTS/02-protect.md` with heading:

   ## Dependency Scan

   Include sections:
   - **Dependency Inventory** — all direct and transitive dependencies with versions
   - **Vulnerability Report** — CVEs found with CVSS scores and affected versions
   - **License Audit** — license per dependency with risk flags
   - **Remediation Plan** — prioritized actions (patch, upgrade, replace) per vulnerable dependency
   - **Dependency Policy** — rules for ongoing dependency management

## Output

The dependency scan written to `.metapowers/security/$ARGUMENTS/02-protect.md`. Present a summary to the user highlighting:
- Total dependencies scanned (direct and transitive)
- Vulnerabilities found by severity (critical, high, medium, low)
- License risks identified
- Top priority remediations and estimated effort

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.