@mitre/hdf-validators

JSON Schema validation for Heimdall Data Format (HDF) documents. Validates all 8 HDF document types against their official schemas to ensure structural correctness and data integrity.

Scope and Responsibilities

hdf-validators provides schema-based validation:

• Validate all HDF document types: Results, Baseline, System, Plan, Amendments, Evidence Package, Comparison, Requirement Change Event
• Auto-detect document type via the generic validate() function
• Detailed error reporting with field-level validation messages
• Support for both TypeScript and Go implementations

hdf-validators vs. hdf-utilities

hdf-validators	hdf-utilities
HDF schema validation	Format syntax validation
"Is this valid HDF?"	"Is this valid JSON/XML/CSV?"
"Does baselines exist and is it an array?"	"Can this string be parsed as JSON?"
Validates structure and types	Validates syntax only
HDF-specific semantic rules	Generic format handling

Example:

• isValidJSON('{"foo": "bar"}') → true (utilities - just checks JSON syntax)
• validateResults({foo: 'bar'}) → false, missing baselines field (validators - checks HDF schema)

Installation

npm install @mitre/hdf-validators

Usage

TypeScript

import { validateResults, validateBaseline } from '@mitre/hdf-validators';

// Validate HDF Results
const hdfResults = {
  baselines: [{
    name: 'My Baseline',
    checksum: { algorithm: 'sha256', value: 'abc123' },
    requirements: [{
      id: 'REQ-001',
      descriptions: [{ label: 'default', data: 'Test requirement' }],
      impact: 0.7,
      tags: { nist: ['AC-1'] },
      results: [{
        status: 'passed',
        codeDesc: 'Control check',
        startTime: '2025-01-01T00:00:00Z'
      }]
    }]
  }],
  targets: [],
  statistics: {}
};

const result = validateResults(hdfResults);

if (result.valid) {
  console.log('✓ Valid HDF Results document');
} else {
  console.error('✗ Validation failed:');
  console.error(result.getErrorMessage());

  // Or access individual errors
  result.errors.forEach(error => {
    console.error(`  ${error.field}: ${error.message}`);
  });
}

// Validate HDF Baseline
const hdfBaseline = {
  name: 'Security Baseline',
  title: 'Example Security Baseline',
  version: '1.0.0',
  checksum: { algorithm: 'sha256', value: 'def456' },
  requirements: [{
    id: 'REQ-001',
    title: 'Access Control',
    descriptions: [{ label: 'default', data: 'Requirement description' }],
    impact: 0.7,
    tags: { nist: ['AC-1', 'AC-2'] }
  }]
};

const baselineResult = validateBaseline(hdfBaseline);

if (!baselineResult.valid) {
  console.error('Validation errors:', baselineResult.errors);
}

// Auto-detect document type
import { validate } from '@mitre/hdf-validators';

const autoResult = validate(someHdfDocument);
// Determines the document type — Results, Baseline, System, Plan, Amendments,
// Comparison or Evidence Package — and validates against that schema

Go

package main

import (
	"encoding/json"
	"fmt"
	"os"

	validators "github.com/mitre/hdf-libs/hdf-validators/go/v3"
)

func main() {
	// Read HDF file
	data, err := os.ReadFile("results.json")
	if err != nil {
		panic(err)
	}

	// Validate HDF Results
	result := validators.ValidateResults(data)

	if result.Valid {
		fmt.Println("✓ Valid HDF Results document")
	} else {
		fmt.Println("✗ Validation failed:")
		fmt.Println(result.Error())

		// Access individual errors
		for _, e := range result.Errors {
			fmt.Printf("  %s: %s\n", e.Field, e.Description)
		}
		os.Exit(1)
	}
}

// Validate HDF Baseline
result := validators.ValidateBaseline(baselineData)

// Use custom schema directory (for development/testing)
validators.SetSchemaDir("./custom-schemas")
result = validators.ValidateResults(data) // Will use schemas from custom directory

// Reset to embedded schemas
validators.SetSchemaDir("")

API Reference

TypeScript

validateResults(data: unknown): ValidationResult

Validate data against the HDF Results schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validateBaseline(data: unknown): ValidationResult

Validate data against the HDF Baseline schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validateComparison(data: unknown): ValidationResult

Validate data against the HDF Comparison schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validateSystem(data: unknown): ValidationResult

Validate data against the HDF System schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validatePlan(data: unknown): ValidationResult

Validate data against the HDF Plan schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validateAmendments(data: unknown): ValidationResult

Validate data against the HDF Amendments schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validateEvidencePackage(data: unknown): ValidationResult

Validate data against the HDF Evidence Package schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validateRequirementChangeEvent(data: unknown): ValidationResult

Validate data against the HDF Requirement Change Event schema.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

validate(data: unknown): ValidationResult

Auto-detect document type and validate.

• Parameters:
  • data - JavaScript object to validate
• Returns: ValidationResult with validation status and errors

ValidationResult

interface ValidationResult {
  valid: boolean;                    // True if validation passed
  errors: ValidationError[];         // Array of validation errors (empty if valid)
  getErrorMessage(): string;         // Formatted error message
}

ValidationError

interface ValidationError {
  field: string;      // JSON path to the field with error
  message: string;    // Description of the validation error
  value?: unknown;    // The invalid value (optional)
}

Go

ValidateResults(data []byte) ValidationResult

Validate JSON bytes against the HDF Results schema.

• Parameters:
  • data - JSON bytes to validate
• Returns: ValidationResult with validation status and errors

ValidateBaseline(data []byte) ValidationResult

Validate JSON bytes against the HDF Baseline schema.

• Parameters:
  • data - JSON bytes to validate
• Returns: ValidationResult with validation status and errors

Validate(data []byte, schemaType SchemaType) ValidationResult

Validate JSON bytes against specified schema type.

• Parameters:
  • data - JSON bytes to validate
  • schemaType - one of TypeResults, TypeBaseline, TypeComparison, TypeSystem, TypePlan, TypeAmendments, TypeEvidencePackage, TypeRequirementChangeEvent
• Returns: ValidationResult with validation status and errors

SetSchemaDir(dir string)

Configure package to load schemas from a directory instead of embedded schemas.

• Parameters:
  • dir - Directory path (empty string to revert to embedded schemas)

GetSchemaDir() string

Get the current schema directory (empty if using embedded schemas).

• Returns: Current schema directory path

ValidationResult

type ValidationResult struct {
    Valid  bool                `json:"valid"`    // True if validation passed
    Errors []ValidationError   `json:"errors"`   // Validation errors (empty if valid)
}

func (r ValidationResult) Error() string  // Formatted error message

ValidationError

type ValidationError struct {
    Field       string `json:"field"`        // JSON path to invalid field
    Description string `json:"description"`  // Error description
    Value       any    `json:"value"`        // Invalid value (optional)
}

Common Validation Errors

Missing Required Fields

baselines: is required

HDF Results must have a baselines array.

Invalid Field Types

baselines: must be array
baselines[0].name: is required

The name field is required for each baseline.

Invalid Enum Values

results[0].status: must be equal to one of the allowed values

Result status must be one of: passed, failed, error, notApplicable, notReviewed.

Invalid Numeric Ranges

impact: must be >= 0 and <= 1

Impact scores must be between 0.0 and 1.0.

Use Cases

Converter Output Validation

Validate that your converter produces valid HDF:

import { convertNessusToHdf } from '@mitre/hdf-converters';
import { validateResults } from '@mitre/hdf-validators';

const hdf = convertNessusToHdf(nessusXml);
const result = validateResults(JSON.parse(hdf));

if (!result.valid) {
  throw new Error(`Invalid HDF output: ${result.getErrorMessage()}`);
}

Pre-Upload Validation

Validate HDF before uploading to Heimdall:

const hdfData = JSON.parse(fs.readFileSync('scan-results.json', 'utf-8'));
const result = validateResults(hdfData);

if (result.valid) {
  uploadToHeimdall(hdfData);
} else {
  console.error('Cannot upload invalid HDF:', result.getErrorMessage());
}

CI/CD Pipeline Validation

# Validate HDF file in CI
hdf validate results.json

# Exit code 0 if valid, non-zero if invalid
if hdf validate scan.json --quiet; then
  echo "✓ HDF validation passed"
else
  echo "✗ HDF validation failed"
  exit 1
fi

Development

# Install dependencies
pnpm install

# Run TypeScript tests
pnpm test:ts

# Run Go tests
pnpm test:go

# Run all tests
pnpm test

# Run tests with coverage
pnpm test:coverage

# Build TypeScript package
pnpm build

# Lint code
pnpm lint

Test Coverage

Both TypeScript and Go implementations maintain >95% test coverage with comprehensive validation tests covering:

• Valid HDF Results and Baseline documents
• Invalid documents (missing fields, wrong types, invalid values)
• Error message formatting
• Custom schema directory loading (Go)
• Integration tests with real HDF fixtures

License

Apache-2.0 © MITRE Corporation