---

name: time-aware-dependency-cve-scanner description: "Scan repositories for newly disclosed CVEs in dependencies after a specific cutoff date. Takes a repository path, cutoff date (YYYY-MM-DD), and optional parameters for transitive dependencies. Parses dependency manifests (package.json, pom.xml, requirements.txt, go.mod, Cargo.toml) and lockfiles to extract exact versions. Queries vulnerability databases (OSV.dev, NVD, GitHub Advisory) to identify CVEs disclosed strictly after the cutoff date. Distinguishes between newly disclosed CVEs and previously known CVEs. Use when: (1) Performing security audits to find new vulnerabilities since last review, (2) Checking if new CVEs affect a historical codebase version, (3) Generating compliance reports showing vulnerability status at specific dates, (4) Tracking security posture changes over time. Supports npm, Maven, pip, Go modules, Cargo, and other major ecosystems."

Time-Aware Dependency CVE Scanner

Scan repositories for newly disclosed CVEs affecting dependencies after a specific cutoff date. This skill helps track when vulnerabilities were introduced and distinguish between pre-existing and newly disclosed security issues.

Quick Start

Basic scan:

python scripts/scan_repository.py /path/to/repo 2023-01-01

Scan only direct dependencies:

python scripts/scan_repository.py /path/to/repo 2023-01-01 --no-transitive

Output as JSON:

python scripts/scan_repository.py /path/to/repo 2023-01-01 --json > report.json

Workflow

1. Parse Dependencies

The scanner automatically detects and parses dependency manifests:

Supported ecosystems:

• npm: package.json, package-lock.json, yarn.lock
• Maven: pom.xml
• Python: requirements.txt, Pipfile.lock, poetry.lock
• Go: go.mod, go.sum
• Cargo: Cargo.toml, Cargo.lock

Manual parsing (if needed):

python scripts/parse_dependencies.py /path/to/repo

This extracts:

• Package names and exact versions
• Direct vs transitive dependency classification
• Ecosystem identification

For detailed manifest formats, see references/dependency_formats.md

2. Query Vulnerability Databases

The scanner queries multiple databases to find CVEs:

Primary source: OSV.dev (Open Source Vulnerabilities)

• No authentication required
• Broad ecosystem coverage (npm, PyPI, Maven, Go, crates.io, etc.)
• Built-in version matching
• Real-time updates

Additional sources:

• NVD (National Vulnerability Database) - Official CVE records
• GitHub Security Advisory - GitHub-curated vulnerabilities

Manual CVE query (for testing):

python scripts/query_cves.py lodash 4.17.20 npm 2023-01-01

For database details and API usage, see references/vulnerability_databases.md

3. Filter by Cutoff Date

The scanner filters CVEs to include only those disclosed after the cutoff date:

• Uses the published date from vulnerability databases
• Excludes CVEs disclosed before or on the cutoff date
• Distinguishes newly disclosed vulnerabilities from pre-existing ones

Example:

• Cutoff date: 2023-01-01
• CVE-2023-12345 published: 2023-06-15 → Included ✓
• CVE-2022-98765 published: 2022-11-20 → Excluded ✗

4. Generate Report

The scanner produces a comprehensive report with:

Summary statistics:

• Total dependencies (direct vs transitive)
• Number of new CVEs found
• CVEs affecting direct vs transitive dependencies
• Severity breakdown (CRITICAL, HIGH, MEDIUM, LOW)

Detailed CVE list: For each CVE:

• CVE identifier (CVE-XXXX-XXXXX or GHSA-XXXX-XXXX-XXXX)
• Affected package and ecosystem
• Version range affected
• Severity score
• Disclosure date
• Summary description

Clear status: If no new CVEs found, explicitly reports "dependency set is clear since the given date"

Use Cases

Security Audit

Scenario: Periodic security review to find vulnerabilities disclosed since last audit

# Last audit was on 2023-06-01, check for new CVEs since then
python scripts/scan_repository.py /path/to/repo 2023-06-01

Output: List of all CVEs disclosed after June 1, 2023 that affect your dependencies

Regression Testing

Scenario: Check if new CVEs affect a specific historical codebase version

# Check if any CVEs disclosed after 2023-01-01 affect code from that date
git checkout <commit-from-2023-01-01>
python scripts/scan_repository.py . 2023-01-01

Output: Shows which vulnerabilities were discovered after the code was written

Compliance Reporting

Scenario: Generate reports showing vulnerability status at specific dates

# Generate quarterly reports
python scripts/scan_repository.py /path/to/repo 2023-01-01 --json > q1_report.json
python scripts/scan_repository.py /path/to/repo 2023-04-01 --json > q2_report.json
python scripts/scan_repository.py /path/to/repo 2023-07-01 --json > q3_report.json

Output: Time-series data showing when vulnerabilities were disclosed

Tracking Security Posture

Scenario: Monitor how security posture changes over time

# Compare vulnerability counts at different dates
python scripts/scan_repository.py /path/to/repo 2022-01-01 | grep "new CVE"
python scripts/scan_repository.py /path/to/repo 2023-01-01 | grep "new CVE"
python scripts/scan_repository.py /path/to/repo 2024-01-01 | grep "new CVE"

Output: Trend analysis of vulnerability accumulation

Advanced Options

Limit Scan Scope

For large repositories, limit the number of dependencies scanned:

python scripts/scan_repository.py /path/to/repo 2023-01-01 --max-deps 50

Direct Dependencies Only

Skip transitive dependencies to focus on direct dependencies:

python scripts/scan_repository.py /path/to/repo 2023-01-01 --no-transitive

JSON Output for Automation

Output structured JSON for integration with other tools:

python scripts/scan_repository.py /path/to/repo 2023-01-01 --json | jq '.summary'

Understanding Results

Report Structure

TIME-AWARE DEPENDENCY CVE SCAN REPORT
======================================================================
Repository: /path/to/repo
Cutoff Date: 2023-01-01
Scan Time: 2024-02-19T10:30:00

DEPENDENCY SUMMARY
----------------------------------------------------------------------
  Total Dependencies: 150
    - Direct: 25
    - Transitive: 125

CVE SUMMARY
----------------------------------------------------------------------
  ⚠ 5 new CVE(s) found after 2023-01-01
    - Affecting direct dependencies: 2
    - Affecting transitive dependencies: 3

  Severity Breakdown:
    - CRITICAL: 1
    - HIGH: 2
    - MEDIUM: 2

DETAILED CVE LIST
----------------------------------------------------------------------

CVE-2023-12345 [CRITICAL]
  Package: lodash (npm)
  Disclosed: 2023-06-15
  Affected Versions: >=4.0.0, <4.17.21
  Summary: Prototype pollution vulnerability...

Interpreting Severity

• CRITICAL: Immediate action required, actively exploited
• HIGH: Serious vulnerability, patch soon
• MEDIUM: Moderate risk, plan remediation
• LOW: Minor issue, low priority
• UNKNOWN: Severity not yet assessed

Next Steps After Scan

1. Review CVEs: Examine each vulnerability's details
2. Check exploitability: Determine if your code uses affected functionality
3. Update dependencies: Upgrade to patched versions
4. Re-scan: Verify fixes with another scan
5. Document: Record findings and remediation actions

Troubleshooting

No dependencies found

• Ensure you're in the repository root
• Check that manifest files exist (package.json, pom.xml, etc.)
• Verify file permissions

API rate limits

• OSV.dev has generous limits, but add delays if hitting limits
• For NVD, get an API key: https://nvd.nist.gov/developers/request-an-api-key
• For GitHub Advisory, set GITHUB_TOKEN environment variable

Parsing errors

• Ensure manifest files are valid JSON/XML/TOML
• Check for syntax errors in dependency declarations
• Some ecosystems may require additional tools (e.g., tomli for Python TOML files)

Dependencies

The scanner scripts require:

• Python 3.7+
• requests library: pip install requests
• Optional: tomli for TOML parsing: pip install tomli