Supported Scanners

Supported Scanners

SPECTRA processes output from the following scanner types. Auto-detection is attempted by default; use --scanner to force a specific parser when auto-detection does not apply.

Trivy

Trivy is an open-source vulnerability scanner for containers, filesystems, Git repositories, and cloud infrastructure.

Auto-detected: Yes — from JSON structure
Input format: JSON (-f json)

Generate scan output:

# Container image scan
trivy image your-image:latest -f json -o trivy.json

# Filesystem scan
trivy fs . -f json -o trivy-fs.json

# Kubernetes scan
trivy k8s --report summary -f json -o trivy-k8s.json

Analyze with SPECTRA:

spectra analyze trivy.json --format both --output reports/container-scan

What SPECTRA extracts from Trivy output:

• CVE identifiers, CVSS scores, and severity ratings
• Affected package names and installed versions
• Fixed versions (when available)
• Vulnerability descriptions for AI contextualization

Semgrep

Semgrep is a static analysis tool for finding bugs, security vulnerabilities, and code quality issues.

Auto-detected: Yes — from JSON structure
Input format: JSON (--json)

Generate scan output:

# Run with default security ruleset
semgrep --config=auto --json > semgrep.json

# Run with a specific ruleset
semgrep --config=p/owasp-top-ten --json > semgrep-owasp.json

Analyze with SPECTRA:

# From file
spectra analyze semgrep.json --format both --output reports/sast-scan

# From stdin (useful in CI pipelines)
semgrep --config=auto --json | spectra analyze --scanner semgrep --format json --output reports/pr-check

What SPECTRA extracts from Semgrep output:

• Rule IDs and CWE mappings
• File paths and line numbers for each finding
• Severity levels and confidence ratings
• Code snippets for contextual analysis

Generic / Nessus / OpenVAS / Burp Suite

The generic scanner mode accepts unstructured text-based output from tools that do not export a parseable JSON format, including:

• Nessus exports
• OpenVAS reports
• Burp Suite HTML or text exports
• Pentest notes and manual findings

Auto-detected: No — must specify --scanner generic
Input format: Plain text or HTML

Analyze with SPECTRA:

# Nessus text export
spectra analyze nessus_report.txt --scanner generic --format both --output reports/nessus

# Burp Suite export
spectra analyze burp_export.txt --scanner generic --format markdown --output reports/burp

# Pentest notes
spectra analyze pentest_notes.txt --scanner generic --format both --output reports/engagement-1

Guidance for generic input quality:

The AI analysis quality scales with input structure. For best results with generic input:

• Include CVE identifiers where available
• Note host/IP context for each finding
• List affected service or component names
• Include severity ratings from the source tool

Adding New Scanners

SPECTRA is open to contributions that add parsers for additional scanner formats. See Contributing for the contribution process and the project repository for the parsers/ module structure.

Scanner Comparison

Next: Output Formats →