4 Integration Methods
Claude Code Skills, Go SDK, CLI, and MCP — pick the one that fits your workflow, from natural language to batch scripting.
Compare methods →
Parse, score, validate, compare & build vulnerability vectors — through Go SDK, CLI, Claude Code Skills & MCP.
CVSS is the industry standard for rating vulnerability severity, but working with vectors programmatically is painful — parsing is error-prone, scoring involves version-specific formulas, comparison is manual, and validation is scattered.
CVSS Skills solves all of this with a single, well-tested toolkit.
Every integration method — Skills, Go SDK, CLI, and MCP — is a thin surface over the same well-tested core packages. Learn the model once; use it everywhere.
The canonical pipeline: a raw vector string is parsed into a typed struct, validated, then run through the version-aware calculator to produce a score and severity rating.
A CVSS vector consists of up to 3 layers of metrics:
| Rating | Score Range | Color |
|---|---|---|
| None | 0.0 | Gray |
| Low | 0.1 – 3.9 | Green |
| Medium | 4.0 – 6.9 | Yellow |
| High | 7.0 – 8.9 | Orange |
| Critical | 9.0 – 10.0 | Red |
A numeric base score maps to exactly one rating band via GetSeverity():
v3.0 vs v3.1 — the score can differ for the same vector
The band boundaries above are identical across versions, but the underlying formulas are not. CVSS v3.1 introduced a defined roundup function and re-weighted a few values (e.g. UI:R = 0.62 in v3.1 vs 0.56 in v3.0). The toolkit reads the CVSS:3.0 / CVSS:3.1 prefix and applies the matching formula automatically — so always keep the version prefix on your vectors.
claude mcp add --scope user cvss-skills -- https://github.com/scagogogo/cvss-skillsgo get github.com/scagogogo/cvss-skills@latestos=$(uname -s | tr '[:upper:]' '[:lower:]'); arch=$(uname -m)
case "$arch" in arm64) arch=aarch64 ;; amd64) arch=x86_64 ;; esac
ver=$(curl -sL https://api.github.com/repos/scagogogo/cvss-skills/releases/latest | sed -nE 's/.*"tag_name":\s*"v?([^"]+)".*/\1/p')
curl -sL "https://github.com/scagogogo/cvss-skills/releases/download/v${ver}/cvss-skills_${ver}_${os}_${arch}.tar.gz" | tar xz
sudo mv cvss /usr/local/bin/go install github.com/scagogogo/cvss-skills/cmd/cvss-cli@latest# Score a vector — works the same in every integration
cvss score "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H"
# Output: 9.8 (Critical)Pick a track and go deep — every module, command, and metric has its own page.
| 🧭 Track | 📄 What you'll learn | 🔗 Start here |
|---|---|---|
| 🎓 Tutorials | Step-by-step lessons: first vector, scoring, validation, batch scripting | Tutorials → |
| 🍳 Recipes | Copy-paste solutions: filter critical vulns, sort, CSV, DB storage | Recipes → |
| 💻 CLI Commands | All 30+ commands, each with flags, real output & JSON examples | CLI Reference → |
| 📦 Go SDK | Core packages: cvss, parser, vector, mock + builder, options, distance, impact | Go SDK → |
| 📐 CVSS Metrics | Every metric dimension — values, scores, version quirks (AV…MA) | Metrics → |
| 🧠 Concepts | Scoring formulas, severity bands, v3.0↔v3.1, validation & distance | Concepts → |