⚡ batch score

CVSS v3.0CVSS v3.1text + json

Synopsis

cvss batch score reads one CVSS vector per line from a file (or stdin) and scores them in parallel. Each input line becomes one output record: a score (severity) vector text line, or a JSON object carrying line, vector, score, and severity. It is the bulk-scoring workhorse for triaging a backlog of vectors.

batch is the parent command for batch operations; score is its scoring subcommand (the sibling is batch validate).

How It Works

Input lines are fanned out to a worker pool that parses and scores each vector in parallel, then reassembled into per-line output records preserving line order.

Usage

cvss batch score [file] [flags]

Flags

Flag	Default	Description
--format string	text	output format: text or json
--workers int	4	number of parallel workers
-h, --help	—	help for score

File or stdin

Pass a file path, or use - (or pipe) to read from stdin. Blank lines and lines starting with # are skipped automatically.

Examples

Score a file of vectors

cat > vectors.txt <<'EOF'
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
EOF
cvss batch score vectors.txt

JSON output piped to jq

cvss batch score --format json --workers 8 vectors.txt | jq 'select(.score >= 9.0)'

From stdin

cat vectors.txt | cvss batch score -

Parse errors go to stderr

Lines that fail to parse are reported on stderr as Line N: parse error: ... and are skipped; the remaining valid vectors are still scored. Per-vector scoring errors are emitted as {"line":N,"error":"..."} JSON objects (in JSON mode) or stderr lines (in text mode).

Underlying API

import (
    "github.com/scagogogo/cvss-skills/pkg/cvss"
    "github.com/scagogogo/cvss-skills/pkg/parser"
)

lines := readLines("vectors.txt") // []string, one vector per line

// Parse in parallel (workerCount workers).
parseResults := parser.BatchParse(lines, 8) // []BatchParseResult

var vectors []*cvss.Cvss3x
var validIndices []int
for _, r := range parseResults {
    if r.Error != nil {
        continue
    }
    vectors = append(vectors, r.Vector)
    validIndices = append(validIndices, r.Index)
}

// Score in parallel.
scoreResults := cvss.BatchScore(vectors, 8) // []BatchScoreResult
for i, r := range scoreResults {
    fmt.Printf("%.1f (%s)  %s\n", r.Score, r.Severity, r.Vector.String())
}

parser.BatchParse(vectors []string, workerCount int) []BatchParseResult parses the lines in parallel; each result carries Index, Vector, and Error. cvss.BatchScore(vectors []*Cvss3x, workerCount int) []BatchScoreResult scores the parsed vectors in parallel; each result carries Vector, Score, Severity, and Error.

Related

• batch validate — validate (not score) many vectors
• score — score a single vector
• sort — sort vectors by score