managing-grype

name	managing-grype
description	Use when working with Grype — anchore Grype vulnerability scanner for container images, filesystems, and SBOMs. Covers image scanning, SBOM-based vulnerability matching, severity filtering, fix tracking, and database management. Use when scanning container images for vulnerabilities, analyzing SBOM security, reviewing vulnerability findings, or managing Grype database updates and scan configurations.
connection_type	grype
preload	false

Grype Management Skill

Manage and analyze Grype vulnerability scans for container images, filesystems, and SBOMs.

CLI Conventions

Authentication

No authentication required for local CLI. For private registries, use standard Docker credentials.

CLI Tool

grype for local scanning with JSON output

Core Helper Function

#!/bin/bash

grype_scan() {
    local target="$1"
    local extra_args="${2:-}"
    grype "$target" -o json $extra_args 2>/dev/null
}

Output Rules

• TOKEN EFFICIENCY: Extract only needed fields with jq
• Target <=50 lines per script output
• Never dump full scan output

Discovery Phase

#!/bin/bash
echo "=== Grype Version ==="
grype version 2>&1 | head -3

echo ""
echo "=== Database Status ==="
grype db status 2>&1

echo ""
echo "=== Supported Sources ==="
echo "Container images (Docker, OCI)"
echo "Filesystem directories"
echo "SBOM files (CycloneDX, SPDX, Syft JSON)"
echo "Archive files (tar, jar, war, zip)"

Analysis Phase

Image Vulnerability Scan

#!/bin/bash
IMAGE="${1:?Container image required (e.g., nginx:latest)}"

echo "=== Scanning: $IMAGE ==="
RESULTS=$(grype_scan "$IMAGE")

echo "$RESULTS" | jq '{
    total: (.matches | length),
    critical: ([.matches[] | select(.vulnerability.severity == "Critical")] | length),
    high: ([.matches[] | select(.vulnerability.severity == "High")] | length),
    medium: ([.matches[] | select(.vulnerability.severity == "Medium")] | length),
    low: ([.matches[] | select(.vulnerability.severity == "Low")] | length),
    fixable: ([.matches[] | select(.vulnerability.fix.state == "fixed")] | length)
}'

echo ""
echo "=== Critical/High Vulnerabilities ==="
echo "$RESULTS" | jq -r '.matches[] | select(.vulnerability.severity == "Critical" or .vulnerability.severity == "High") | "\(.vulnerability.severity)\t\(.vulnerability.id)\t\(.artifact.name[0:25])\tv\(.artifact.version)\tfix:\(.vulnerability.fix.versions[0] // "none")"' \
    | column -t | head -20

Fix Prioritization

#!/bin/bash
IMAGE="${1:?Container image required}"

echo "=== Fixable Vulnerabilities ==="
grype_scan "$IMAGE" \
    | jq -r '.matches[] | select(.vulnerability.fix.state == "fixed") | "\(.vulnerability.severity)\t\(.vulnerability.id)\t\(.artifact.name[0:25])\tv\(.artifact.version) -> \(.vulnerability.fix.versions[0])"' \
    | sort | column -t | head -20

echo ""
echo "=== Unfixed Critical/High ==="
grype_scan "$IMAGE" \
    | jq -r '.matches[] | select(.vulnerability.fix.state != "fixed" and (.vulnerability.severity == "Critical" or .vulnerability.severity == "High")) | "\(.vulnerability.severity)\t\(.vulnerability.id)\t\(.artifact.name[0:25])\tv\(.artifact.version)"' \
    | column -t | head -15

Multi-Image Comparison

#!/bin/bash
echo "=== Multi-Image Vulnerability Comparison ==="
for IMAGE in "$@"; do
    RESULTS=$(grype_scan "$IMAGE" 2>/dev/null)
    CRITICAL=$(echo "$RESULTS" | jq '[.matches[] | select(.vulnerability.severity == "Critical")] | length')
    HIGH=$(echo "$RESULTS" | jq '[.matches[] | select(.vulnerability.severity == "High")] | length')
    TOTAL=$(echo "$RESULTS" | jq '.matches | length')
    FIXABLE=$(echo "$RESULTS" | jq '[.matches[] | select(.vulnerability.fix.state == "fixed")] | length')
    echo "$IMAGE\ttotal:${TOTAL}\tcritical:${CRITICAL}\thigh:${HIGH}\tfixable:${FIXABLE}"
done | column -t

SBOM Scanning

#!/bin/bash
SBOM_FILE="${1:?SBOM file path required}"

echo "=== SBOM Vulnerability Scan ==="
RESULTS=$(grype "sbom:${SBOM_FILE}" -o json 2>/dev/null)

echo "$RESULTS" | jq '{
    total_vulns: (.matches | length),
    by_severity: (.matches | group_by(.vulnerability.severity) | map({severity: .[0].vulnerability.severity, count: length}) | sort_by(.count) | reverse)
}'

echo ""
echo "=== Top Vulnerable Components ==="
echo "$RESULTS" | jq -r '[.matches[] | {name: .artifact.name, severity: .vulnerability.severity}] | group_by(.name) | map({name: .[0].name, count: length, worst: ([.[].severity] | if any(. == "Critical") then "Critical" elif any(. == "High") then "High" else "Medium" end)}) | sort_by(.count) | reverse | .[:10][] | "\(.count)\t\(.worst)\t\(.name)"' \
    | column -t

Output Format

Present results as a structured report:

Managing Grype Report
═════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative

Common Pitfalls

• Database freshness: Run grype db update regularly -- stale DB misses recent CVEs
• Image pull: Grype pulls images from registry -- ensure registry credentials are configured
• Output formats: Use -o json for parsing, -o table for display, -o cyclonedx for SBOM
• Severity filtering: Use --only-fixed to show only fixable vulns, --fail-on critical for CI/CD
• Distroless images: May have limited OS-level detection -- rely on language-specific matchers
• SBOM prefix: Use sbom: prefix for SBOM file input, dir: for filesystem scanning