---
name: dependency-track
description: Comprehensive guide for Dependency-Track - Software Composition Analysis (SCA) and SBOM management platform. USE WHEN deploying Dependency-Track, integrating with CI/CD pipelines, configuring vulnerability scanning, managing SBOMs, setting up policy compliance, troubleshooting installation issues, or working with the REST API.
triggers:
  - dependency-track
  - dependencytrack
  - dtrack
  - sbom
  - software bill of materials
  - software composition analysis
  - sca platform
  - vulnerability scanning
  - component analysis
  - cyclonedx
  - bom upload
  - license compliance
  - supply chain security
  - hyades
---

# Dependency-Track Skill

Comprehensive guide for implementing, deploying, and operating **Dependency-Track** - an intelligent Software Composition Analysis (SCA) platform that identifies and reduces risk in the software supply chain through SBOM management.

**Current Versions:**
- Helm Chart: `0.40.0`
- App Version: `4.13.6`
- Helm Repository: `https://dependencytrack.github.io/helm-charts`

## Overview

Dependency-Track is an API-first platform that:
- Consumes and produces CycloneDX SBOMs and VEX documents
- Monitors components for known vulnerabilities across the entire portfolio
- Integrates with NVD, GitHub Advisories, OSS Index, Snyk, Trivy, OSV, and VulnDB
- Provides policy enforcement for security, license, and operational compliance
- Supports OAuth 2.0, OIDC, LDAP, Active Directory authentication
- Supports EPSS (Exploit Prediction Scoring System) for prioritization
- Identifies APIs and external service components

## Quick Reference

| Resource | Path |
|----------|------|
| Deployment Templates | `references/deployment/` |
| CI/CD Integration | `references/cicd/` |
| API Examples | `references/api/` |
| Policy Templates | `references/policies/` |
| Troubleshooting | `references/troubleshooting.md` |

---

## 1. Deployment Options

### Docker Compose (Recommended for Production)

```bash
# Download official docker-compose
curl -LO https://dependencytrack.org/docker-compose.yml

# Start services
docker compose up -d

# Access UI at http://localhost:8080
# Default credentials: admin / admin
```

**Minimum Requirements:**
- API Server: 4.5GB RAM, 2 CPU cores
- Frontend: 512MB RAM, 1 CPU core

**Recommended Requirements:**
- API Server: 16GB RAM, 4 CPU cores
- Frontend: 1GB RAM, 2 CPU cores

### Kubernetes with Helm

```bash
# Add Helm repository
helm repo add dependency-track https://dependencytrack.github.io/helm-charts
helm repo update

# View available charts
helm search repo dependency-track

# Install with custom values
helm install dtrack dependency-track/dependency-track \
    --namespace dtrack \
    --create-namespace \
    -f values-production.yaml

# Upgrade existing installation
helm upgrade dtrack dependency-track/dependency-track \
    --namespace dtrack \
    -f values-production.yaml
```

**Available Charts:**
| Chart | Description | Status |
|-------|-------------|--------|
| `dependency-track/dependency-track` | Monolithic deployment (v4.x) | Production Ready |
| `dependency-track/hyades` | Microservices deployment (v5.x) | Incubating (Not GA) |

See `references/deployment/` for complete manifests and Helm values.

### ArgoCD GitOps Deployment (AKS)

For enterprise Kubernetes deployments using ArgoCD with multi-source ApplicationSets:

**Directory Structure:**
```
infra-team/
  applicationset/
    dependency-track.yaml       # ArgoCD ApplicationSet definition
argo-cd-helm-values/
  kube-addons/
    dependency-track/
      azure-ad-setup.sh         # Azure AD App Registration script
      cafehyna-dev/
        values.yaml             # Environment-specific Helm values
        secretproviderclass.yaml # Azure Key Vault CSI integration
```

**ApplicationSet Example (Multi-Source):**
```yaml
apiVersion: argoproj.io/v1alpha1
kind: ApplicationSet
metadata:
  name: dependency-track
  namespace: argocd
spec:
  generators:
    - list:
        elements:
          - cluster: cafehyna-dev
            url: https://aks-cluster-url:443
            project: kube-addons
            branch: main
            environment: development
            urlSuffix: ".dev"
  template:
    metadata:
      name: "{{cluster}}-dependency-track"
      finalizers:
        - resources-finalizer.argocd.argoproj.io
    spec:
      project: "{{project}}"
      sources:
        # Source 1: Helm chart from official repository
        - chart: dependency-track
          repoURL: https://dependencytrack.github.io/helm-charts
          targetRevision: "0.40.0"
          helm:
            releaseName: dependency-track
            valueFiles:
              - $values/argo-cd-helm-values/kube-addons/dependency-track/{{cluster}}/values.yaml
        # Source 2: Values repository reference
        - repoURL: https://your-git-repo.git
          targetRevision: "{{branch}}"
          ref: values
        # Source 3: Additional manifests (SecretProviderClass)
        - repoURL: https://your-git-repo.git
          targetRevision: "{{branch}}"
          path: argo-cd-helm-values/kube-addons/dependency-track/{{cluster}}
          directory:
            exclude: values.yaml
      destination:
        server: "{{url}}"
        namespace: dependency-track
      syncPolicy:
        automated:
          prune: true
          selfHeal: true
        syncOptions:
          - ServerSideApply=true
          - CreateNamespace=true
```

**Azure Key Vault CSI Driver (SecretProviderClass):**
```yaml
apiVersion: secrets-store.csi.x-k8s.io/v1
kind: SecretProviderClass
metadata:
  name: dependency-track-secrets-kv
  namespace: dependency-track
spec:
  provider: azure
  parameters:
    usePodIdentity: "false"
    useVMManagedIdentity: "true"
    userAssignedIdentityID: "<managed-identity-client-id>"
    keyvaultName: "<key-vault-name>"
    tenantId: "<tenant-id>"
    objects: |
      array:
        - |
          objectName: "dtrack-oidc-client-id"
          objectType: "secret"
          objectAlias: "oidc-client-id"
        - |
          objectName: "dtrack-oidc-client-secret"
          objectType: "secret"
          objectAlias: "oidc-client-secret"
  secretObjects:
    - secretName: dependency-track-secrets
      type: Opaque
      data:
        - objectName: oidc-client-id
          key: oidc-client-id
        - objectName: oidc-client-secret
          key: oidc-client-secret
```

**AKS Spot Instance Tolerations (Cost Optimization):**
```yaml
apiServer:
  tolerations:
    - key: kubernetes.azure.com/scalesetpriority
      operator: Equal
      value: spot
      effect: NoSchedule
  affinity:
    nodeAffinity:
      preferredDuringSchedulingIgnoredDuringExecution:
        - weight: 100
          preference:
            matchExpressions:
              - key: kubernetes.azure.com/scalesetpriority
                operator: In
                values:
                  - spot
```

**Key ignoreDifferences for ArgoCD:**
```yaml
ignoreDifferences:
  # StatefulSet VolumeClaimTemplates (ServerSideApply drift)
  - group: apps
    kind: StatefulSet
    jqPathExpressions:
      - .spec.volumeClaimTemplates[]?.apiVersion
      - .spec.volumeClaimTemplates[]?.kind
      - .spec.volumeClaimTemplates[]?.status
  # PVC dynamic fields
  - group: ""
    kind: PersistentVolumeClaim
    jsonPointers:
      - /status
      - /spec/volumeName
  # Azure Key Vault synced secrets
  - group: ""
    kind: Secret
    name: dependency-track-secrets
    jsonPointers:
      - /data
```

See `references/deployment/argocd/` for complete ApplicationSet and values examples.

### Hyades (Next-Generation Architecture)

Hyades is the incubating project for Dependency-Track v5, decoupling the monolithic API server into separate, scalable microservices.

**Components:**
- `hyades-apiserver` - Core API server
- `hyades-frontend` - Web UI
- `hyades-notification-publisher` - Notification handling
- `hyades-repository-meta-analyzer` - Repository metadata analysis
- `hyades-vulnerability-analyzer` - Vulnerability scanning

**Requirements:**
- External PostgreSQL database
- Apache Kafka cluster
- Kubernetes 1.19+

```bash
# Install Hyades (NOT PRODUCTION READY)
helm install hyades dependency-track/hyades \
    --namespace dtrack \
    --create-namespace \
    --set common.database.jdbcUrl="jdbc:postgresql://postgres:5432/dtrack" \
    --set common.database.username="dtrack" \
    --set common.database.password="secret" \
    --set common.kafka.bootstrapServers="kafka:9092"
```

> **Warning:** Hyades is NOT generally available. Breaking changes may occur without notice. Use only in test environments. GA roadmap: https://github.com/DependencyTrack/hyades/issues/860

---

## 2. Initial Configuration

### First-Time Setup

1. **Wait for initialization** (10-30+ minutes):
   - Creates default users, teams, permissions
   - Mirrors NVD, GitHub Advisories (do not interrupt)
   - Check progress: `docker logs -f dependency-track-apiserver`

2. **Change default password** immediately after first login

3. **Enable analyzers** in Administration > Analyzers:
   - Internal Analyzer (built-in)
   - OSS Index (requires API token from ossindex.sonatype.org)
   - Snyk (optional, requires token)
   - Trivy (optional)

### Database Configuration (Production)

```yaml
# docker-compose override for PostgreSQL
services:
  apiserver:
    environment:
      ALPINE_DATABASE_MODE: external
      ALPINE_DATABASE_URL: jdbc:postgresql://postgres:5432/dtrack
      ALPINE_DATABASE_DRIVER: org.postgresql.Driver
      ALPINE_DATABASE_USERNAME: dtrack
      ALPINE_DATABASE_PASSWORD: ${DB_PASSWORD}
```

### LDAP/Active Directory

```properties
# Active Directory configuration
ALPINE_LDAP_ENABLED=true
ALPINE_LDAP_SERVER_URL=ldap://ldap.example.com:3268
ALPINE_LDAP_BASEDN=dc=example,dc=com
ALPINE_LDAP_SECURITY_AUTH=simple
ALPINE_LDAP_AUTH_USERNAME_FORMAT=%s@example.com
ALPINE_LDAP_ATTRIBUTE_NAME=userPrincipalName
ALPINE_LDAP_ATTRIBUTE_MAIL=mail
ALPINE_LDAP_GROUPS_FILTER=(&(objectClass=group)(objectCategory=Group))
ALPINE_LDAP_USER_GROUPS_FILTER=(&(objectClass=group)(member:1.2.840.113556.1.4.1941:={USER_DN}))
```

### OpenID Connect (Azure AD, Okta, Keycloak)

```properties
ALPINE_OIDC_ENABLED=true
ALPINE_OIDC_ISSUER=https://login.microsoftonline.com/{tenant}/v2.0
ALPINE_OIDC_CLIENT_ID=your-client-id
ALPINE_OIDC_USERNAME_CLAIM=preferred_username
ALPINE_OIDC_TEAMS_CLAIM=groups
ALPINE_OIDC_USER_PROVISIONING=true
ALPINE_OIDC_TEAM_SYNCHRONIZATION=true
```

### Azure AD OIDC Groups Integration (Detailed Guide)

#### How OIDC Groups Work in Dependency-Track

1. **Group Claim Flow**:
   - User authenticates via Azure AD
   - Azure AD returns ID token with `groups` claim (array of Group Object IDs)
   - Dependency-Track reads the `groups` claim (configured via `ALPINE_OIDC_TEAMS_CLAIM`)
   - Groups are matched to Teams via OpenID Connect Groups mappings

2. **Important Behavior**:
   - Groups only appear in Administration > OpenID Connect Groups **after a user from that group authenticates**
   - Teams must exist before group mapping (auto-created if `ALPINE_OIDC_TEAM_SYNCHRONIZATION=true`)
   - Group Object IDs (UUIDs) are used, not display names

#### Method 1: API-Based Setup (Recommended)

Use the REST API to create teams, OIDC groups, and mappings:

```bash
# Set environment variables
export DTRACK_URL="https://dtrack.example.com"
export DTRACK_API_KEY="your-api-key"

# Azure AD Group Object ID
AZURE_GROUP_ID="31d6daa5-5cc2-4e5f-9bf5-75ee8e09198c"

# 1. Create a Team (if doesn't exist)
curl -X PUT "${DTRACK_URL}/api/v1/team" \
  -H "X-Api-Key: ${DTRACK_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"name": "Administrators"}'

# 2. Get the Team UUID
TEAM_UUID=$(curl -s -H "X-Api-Key: ${DTRACK_API_KEY}" \
  "${DTRACK_URL}/api/v1/team" | jq -r '.[] | select(.name=="Administrators") | .uuid')

# 3. Create OIDC Group (required before mapping)
# NOTE: Use Azure AD Group Object ID as both uuid and name
curl -X PUT "${DTRACK_URL}/api/v1/oidc/group" \
  -H "X-Api-Key: ${DTRACK_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "{\"uuid\": \"${AZURE_GROUP_ID}\", \"name\": \"${AZURE_GROUP_ID}\"}"

# 4. Get the created OIDC Group UUID
OIDC_GROUP_UUID=$(curl -s -H "X-Api-Key: ${DTRACK_API_KEY}" \
  "${DTRACK_URL}/api/v1/oidc/group" | jq -r ".[] | select(.name==\"${AZURE_GROUP_ID}\") | .uuid")

# 5. Create OIDC Group-to-Team mapping
# IMPORTANT: Use UUID strings directly, NOT objects
curl -X PUT "${DTRACK_URL}/api/v1/oidc/mapping" \
  -H "X-Api-Key: ${DTRACK_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "{\"team\": \"${TEAM_UUID}\", \"group\": \"${OIDC_GROUP_UUID}\"}"

# 6. Assign permissions to Team
curl -X POST "${DTRACK_URL}/api/v1/permission/BOM_UPLOAD/team/${TEAM_UUID}" \
  -H "X-Api-Key: ${DTRACK_API_KEY}"
```

> **IMPORTANT:** The `/api/v1/oidc/mapping` endpoint expects UUID **strings**, not objects.
> Using `{"team": {"uuid": "..."}, "group": {"uuid": "..."}}` will return HTTP 400.

#### Verifying OIDC Group Mappings

Use these commands to verify that OIDC groups are properly mapped to teams:

```bash
# Set environment variables
export DTRACK_URL="https://dtrack.example.com"
export DTRACK_API_KEY="your-api-key"

# List all teams with their OIDC group mappings
/usr/bin/curl -s "${DTRACK_URL}/api/v1/team" \
  -H "X-Api-Key: ${DTRACK_API_KEY}" | \
  /usr/bin/jq '[.[] | select(.mappedOidcGroups | length > 0) | {
    team: .name,
    oidcGroups: [.mappedOidcGroups[] | .group.name]
  }]'

# List all OIDC groups registered in D-Track
/usr/bin/curl -s "${DTRACK_URL}/api/v1/oidc/group" \
  -H "X-Api-Key: ${DTRACK_API_KEY}" | \
  /usr/bin/jq '.[] | {uuid: .uuid, name: .name}'

# Check specific team's OIDC mappings
TEAM_NAME="Administrators"
/usr/bin/curl -s "${DTRACK_URL}/api/v1/team" \
  -H "X-Api-Key: ${DTRACK_API_KEY}" | \
  /usr/bin/jq --arg team "$TEAM_NAME" '.[] | select(.name==$team) | {
    name: .name,
    uuid: .uuid,
    oidcGroups: [.mappedOidcGroups[]? | {name: .group.name, uuid: .group.uuid}]
  }'
```

**Example Output (cafehyna-dev cluster):**
```json
[
  {"team": "Administrators", "oidcGroups": ["31d6daa5-5cc2-4e5f-9bf5-75ee8e09198c"]},
  {"team": "Auditors", "oidcGroups": ["39d0ddb2-8a0d-47c9-bc55-1993aadf7fe8"]},
  {"team": "Developers", "oidcGroups": ["7144c503-7934-40fe-99e2-f45f67cb383e"]},
  {"team": "Moderators", "oidcGroups": ["d34a2ea3-34c9-4744-a031-4e4a55d72746"]},
  {"team": "ReadOnly", "oidcGroups": ["7109c419-028e-4b7b-8246-66517d512c3c"]}
]
```

> **Note:** Use full paths (`/usr/bin/curl`, `/usr/bin/jq`) to avoid shell alias issues in automated environments.

#### Method 2: Manual UI Configuration

1. **Create Teams**:
   - Go to Administration > Access Management > Teams
   - Create teams: Administrators, Moderators, Auditors, Developers
   - Assign permissions to each team

2. **Map Groups**:
   - Go to Administration > Access Management > OpenID Connect Groups
   - Click "Create Group"
   - Enter Azure AD Group Object ID (e.g., `31d6daa5-5cc2-4e5f-9bf5-75ee8e09198c`)
   - Select Team to map to
   - Save

#### Retrieving Azure AD Group Object IDs

```bash
# List all Azure AD groups with DependencyTrack in name
az ad group list --filter "startswith(displayName, 'G-Usuarios-DependencyTrack')" \
  --query "[].{name:displayName, id:id}" -o table

# Get specific group Object ID
az ad group show --group "G-Usuarios-DependencyTrack-Admin" --query id -o tsv
```

#### Validating Group Claims in ID Token

1. **Decode JWT Token** (use jwt.io):
   - Login to Dependency-Track
   - Capture network request to `/api/v1/oidc/callback`
   - Decode the `id_token` to verify `groups` claim contains expected UUIDs

2. **Azure AD Token Configuration Checklist**:
   - App Registration has `groupMembershipClaims: SecurityGroup`
   - Optional Claims include `groups` in ID Token
   - API Permissions: `GroupMember.Read.All` (delegated)
   - Admin consent granted

#### Environment Variables Reference

| Variable | Description | Example |
|----------|-------------|---------|
| `ALPINE_OIDC_ENABLED` | Enable OIDC | `true` |
| `ALPINE_OIDC_ISSUER` | Azure AD issuer URL | `https://login.microsoftonline.com/{tenant}/v2.0` |
| `ALPINE_OIDC_CLIENT_ID` | App Registration Client ID | From Key Vault |
| `ALPINE_OIDC_CLIENT_SECRET` | App Registration Secret | From Key Vault |
| `ALPINE_OIDC_USERNAME_CLAIM` | Claim for username | `preferred_username` |
| `ALPINE_OIDC_TEAMS_CLAIM` | Claim for groups | `groups` |
| `ALPINE_OIDC_USER_PROVISIONING` | Auto-create users | `true` |
| `ALPINE_OIDC_TEAM_SYNCHRONIZATION` | Auto-sync team membership | `true` |

---

## 3. CI/CD Integration

### Workflow: BOM Upload Pipeline

```
┌─────────────┐     ┌─────────────┐     ┌──────────────────┐
│   Build     │────▶│ Generate    │────▶│    Upload to     │
│   Project   │     │ CycloneDX   │     │ Dependency-Track │
└─────────────┘     │    SBOM     │     └──────────────────┘
                    └─────────────┘              │
                                                 ▼
                    ┌─────────────┐     ┌──────────────────┐
                    │   Break     │◀────│    Evaluate      │
                    │   Build?    │     │    Policies      │
                    └─────────────┘     └──────────────────┘
```

### Jenkins Pipeline

```groovy
pipeline {
    agent any

    environment {
        DTRACK_URL = 'https://dtrack.example.com'
        DTRACK_API_KEY = credentials('dependency-track-api-key')
    }

    stages {
        stage('Build') {
            steps {
                sh 'mvn clean package'
            }
        }

        stage('Generate SBOM') {
            steps {
                sh 'mvn org.cyclonedx:cyclonedx-maven-plugin:makeAggregateBom'
            }
        }

        stage('Upload SBOM') {
            steps {
                dependencyTrackPublisher(
                    artifact: 'target/bom.xml',
                    projectId: env.PROJECT_UUID,
                    synchronous: true,
                    failedTotalCritical: 0,
                    failedTotalHigh: 5
                )
            }
        }
    }
}
```

### GitHub Actions

```yaml
name: SBOM Security Scan

on:
  push:
    branches: [main]
  pull_request:
    branches: [main]

jobs:
  sbom-scan:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Generate SBOM
        uses: CycloneDX/gh-node-module-generatebom@v1
        with:
          output: ./bom.json

      - name: Upload to Dependency-Track
        uses: DependencyTrack/gh-upload-sbom@v3
        with:
          serverHostname: ${{ secrets.DTRACK_URL }}
          apiKey: ${{ secrets.DTRACK_API_KEY }}
          projectName: ${{ github.repository }}
          projectVersion: ${{ github.ref_name }}
          bomFilename: ./bom.json
          autoCreate: true
```

**Official GitHub Action Options (`DependencyTrack/gh-upload-sbom@v3`):**
| Parameter | Required | Default | Description |
|-----------|----------|---------|-------------|
| `serverHostname` | Yes | - | Dependency-Track server URL |
| `apiKey` | Yes | - | API key with BOM_UPLOAD permission |
| `projectName` | Yes | - | Name of the project |
| `projectVersion` | Yes | - | Version of the project |
| `bomFilename` | Yes | - | Path to CycloneDX SBOM file |
| `autoCreate` | No | `false` | Auto-create project if not exists |
| `parentName` | No | - | Parent project name |
| `parentVersion` | No | - | Parent project version |
| `projectUuid` | No | - | Use UUID instead of name/version |

### GitLab CI

```yaml
stages:
  - build
  - security

generate-sbom:
  stage: build
  image: cyclonedx/cyclonedx-cli
  script:
    - cyclonedx-py -r requirements.txt -o bom.json
  artifacts:
    paths:
      - bom.json

upload-sbom:
  stage: security
  image: curlimages/curl
  script:
    - |
      curl -X PUT "${DTRACK_URL}/api/v1/bom" \
        -H "X-Api-Key: ${DTRACK_API_KEY}" \
        -H "Content-Type: application/json" \
        -d @- << EOF
      {
        "projectName": "${CI_PROJECT_NAME}",
        "projectVersion": "${CI_COMMIT_REF_NAME}",
        "autoCreate": true,
        "bom": "$(base64 -w0 bom.json)"
      }
      EOF
```

### Azure DevOps Pipeline

```yaml
trigger:
  - main

pool:
  vmImage: 'ubuntu-latest'

variables:
  - group: dependency-track-credentials

steps:
  - task: Maven@4
    inputs:
      mavenPomFile: 'pom.xml'
      goals: 'package org.cyclonedx:cyclonedx-maven-plugin:makeAggregateBom'

  - task: Bash@3
    displayName: 'Upload SBOM to Dependency-Track'
    inputs:
      targetType: 'inline'
      script: |
        BOM_CONTENT=$(base64 -w0 target/bom.xml)
        curl -X PUT "$(DTRACK_URL)/api/v1/bom" \
          -H "X-Api-Key: $(DTRACK_API_KEY)" \
          -H "Content-Type: application/json" \
          -d "{\"projectName\":\"$(Build.Repository.Name)\",\"projectVersion\":\"$(Build.SourceBranchName)\",\"autoCreate\":true,\"bom\":\"${BOM_CONTENT}\"}"
```

---

## 4. REST API Usage

### Authentication

Generate API keys in Administration > Access Management > Teams > [Team] > API Keys.

```bash
# All API calls require the X-Api-Key header
curl -H "X-Api-Key: YOUR_API_KEY" \
     https://dtrack.example.com/api/v1/project
```

### OpenAPI Specification

Access at:
- JSON: `http://localhost:8081/api/openapi.json`
- YAML: `http://localhost:8081/api/openapi.yaml`

**Note:** Use port 8081 (API server), not 8080 (frontend).

### Common API Operations

```bash
# List all projects
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/project" | jq

# Get project by name and version
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/project/lookup?name=myapp&version=1.0.0" | jq

# Upload SBOM (Base64 encoded)
curl -X PUT "${DTRACK_URL}/api/v1/bom" \
     -H "X-Api-Key: ${API_KEY}" \
     -H "Content-Type: application/json" \
     -d "{
       \"projectName\": \"my-application\",
       \"projectVersion\": \"1.0.0\",
       \"autoCreate\": true,
       \"bom\": \"$(base64 -w0 bom.json)\"
     }"

# Upload SBOM (multipart - no encoding needed)
curl -X POST "${DTRACK_URL}/api/v1/bom" \
     -H "X-Api-Key: ${API_KEY}" \
     -F "projectName=my-application" \
     -F "projectVersion=1.0.0" \
     -F "autoCreate=true" \
     -F "bom=@bom.json"

# Get vulnerabilities for a project
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/vulnerability/project/${PROJECT_UUID}" | jq

# Get policy violations
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/violation/project/${PROJECT_UUID}" | jq

# Get component dependencies
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/component/project/${PROJECT_UUID}" | jq

# Create project
curl -X PUT "${DTRACK_URL}/api/v1/project" \
     -H "X-Api-Key: ${API_KEY}" \
     -H "Content-Type: application/json" \
     -d '{
       "name": "new-project",
       "version": "1.0.0",
       "description": "Project description",
       "tags": [{"name": "production"}]
     }'
```

---

## 5. Policy Configuration

### Policy Types

| Type | Purpose | Example Conditions |
|------|---------|-------------------|
| **License** | Control allowed/forbidden licenses | Apache-2.0 allowed, GPL-3.0 forbidden |
| **Security** | Vulnerability severity thresholds | No Critical, max 5 High |
| **Operational** | Component allowlists/denylists | Block log4j < 2.17.0 |

### Creating Policies via API

```bash
# Create a security policy
curl -X PUT "${DTRACK_URL}/api/v1/policy" \
     -H "X-Api-Key: ${API_KEY}" \
     -H "Content-Type: application/json" \
     -d '{
       "name": "No Critical Vulnerabilities",
       "operator": "ANY",
       "violationState": "FAIL",
       "policyConditions": [
         {
           "subject": "SEVERITY",
           "operator": "IS",
           "value": "CRITICAL"
         }
       ]
     }'

# Create a license policy (denylist approach)
curl -X PUT "${DTRACK_URL}/api/v1/policy" \
     -H "X-Api-Key: ${API_KEY}" \
     -H "Content-Type: application/json" \
     -d '{
       "name": "No Copyleft Licenses",
       "operator": "ANY",
       "violationState": "WARN",
       "policyConditions": [
         {
           "subject": "LICENSE_GROUP",
           "operator": "IS",
           "value": "Copyleft"
         }
       ]
     }'

# Create operational policy (block specific component)
curl -X PUT "${DTRACK_URL}/api/v1/policy" \
     -H "X-Api-Key: ${API_KEY}" \
     -H "Content-Type: application/json" \
     -d '{
       "name": "Block Vulnerable Log4j",
       "operator": "ALL",
       "violationState": "FAIL",
       "policyConditions": [
         {
           "subject": "COORDINATES",
           "operator": "MATCHES",
           "value": "{\"group\":\"org.apache.logging.log4j\",\"name\":\"log4j-core\",\"version\":\"<2.17.0\"}"
         }
       ]
     }'
```

### Policy Condition Subjects

- `AGE` - Component age
- `COORDINATES` - Group/Name/Version
- `CPE` - Common Platform Enumeration
- `CWE` - Common Weakness Enumeration
- `EPSS` - Exploit Prediction Scoring System
- `LICENSE` - Specific license
- `LICENSE_GROUP` - License category
- `PACKAGE_URL` - Package URL (PURL)
- `SEVERITY` - Vulnerability severity
- `SWID_TAGID` - Software ID tag
- `VERSION` - Component version
- `COMPONENT_HASH` - MD5, SHA, Blake hashes

---

## 6. Notifications & Integrations

### Webhook Configuration

```json
{
  "notification": {
    "level": "INFORMATIONAL",
    "scope": "PORTFOLIO",
    "group": "NEW_VULNERABILITY",
    "timestamp": "2024-01-15T10:30:00Z",
    "subject": {
      "component": {
        "name": "log4j-core",
        "version": "2.14.1"
      },
      "vulnerability": {
        "vulnId": "CVE-2021-44228",
        "severity": "CRITICAL"
      },
      "affectedProjects": [...]
    }
  }
}
```

### Notification Groups

- `NEW_VULNERABILITY` - New vulnerability identified
- `NEW_VULNERABLE_DEPENDENCY` - New vulnerable component added
- `ANALYSIS_DECISION_CHANGE` - Audit decision modified
- `POLICY_VIOLATION` - Policy violation detected
- `BOM_CONSUMED` - SBOM processed
- `BOM_PROCESSED` - SBOM analysis complete
- `VEX_CONSUMED` - VEX document processed

### Integration Targets

- **Slack/Teams**: Direct webhook support
- **Email**: SMTP configuration
- **Jira**: Native integration
- **DefectDojo**: Vulnerability aggregation
- **Fortify SSC**: AppSec platform
- **ThreadFix**: Vulnerability management
- **Kenna Security**: Risk prioritization

---

## 7. SBOM Generation Tools

### By Language/Ecosystem

| Language | Tool | Command |
|----------|------|---------|
| Java/Maven | cyclonedx-maven-plugin | `mvn org.cyclonedx:cyclonedx-maven-plugin:makeAggregateBom` |
| Java/Gradle | cyclonedx-gradle-plugin | `gradle cyclonedxBom` |
| Node.js | @cyclonedx/cyclonedx-npm | `npx @cyclonedx/cyclonedx-npm --output-file bom.json` |
| Python | cyclonedx-py | `cyclonedx-py -r requirements.txt -o bom.json` |
| Go | cyclonedx-gomod | `cyclonedx-gomod mod -json -output bom.json` |
| .NET | CycloneDX | `dotnet CycloneDX <project> -o bom.json` |
| Rust | cargo-cyclonedx | `cargo cyclonedx -f json` |
| Container | Syft | `syft <image> -o cyclonedx-json > bom.json` |

### Universal Tools

```bash
# Syft - multi-ecosystem SBOM generator
syft packages dir:. -o cyclonedx-json > bom.json

# Trivy - with vulnerability scanning
trivy fs --format cyclonedx --output bom.json .

# cdxgen - comprehensive generator
cdxgen -o bom.json
```

---

## 8. Best Practices

### Deployment

1. **Use external PostgreSQL** for production (not embedded H2)
2. **Allocate sufficient RAM** (minimum 4.5GB for API server)
3. **Configure persistent storage** for `/data` directory
4. **Enable TLS** with proper certificates
5. **Set up regular backups** of database and configuration

### Operations

1. **Generate SBOMs in CI/CD** - automate at build time
2. **Enable OSS Index Analyzer** - required for PURL-based scanning
3. **Mirror NVD locally** - improves performance
4. **Use synchronous uploads** in pipelines for immediate feedback
5. **Implement policy gates** - fail builds on violations

### Security

1. **Rotate API keys** regularly
2. **Use team-scoped keys** with minimal permissions
3. **Enable audit logging** for compliance
4. **Configure OIDC/LDAP** for enterprise authentication
5. **Contractually require SBOMs** from vendors

### Performance

1. **Schedule heavy operations** during off-hours
2. **Use pagination** for large API responses
3. **Implement caching** for frequently accessed data
4. **Monitor analyzer queue** depth
5. **Scale horizontally** with multiple API server replicas

---

## 9. Troubleshooting

### Common Issues

| Issue | Cause | Solution |
|-------|-------|----------|
| No vulnerabilities found | OSS Index not enabled | Enable in Admin > Analyzers, get API token |
| Analyzer not running | 6-hour schedule | Re-upload SBOM or wait for scheduled run |
| Container crashes | Insufficient RAM | Allocate minimum 4.5GB to API server |
| 413 Request Entity Too Large | Nginx body size limit | Add `nginx.ingress.kubernetes.io/proxy-body-size: "100m"` |
| PKIX path building error | Self-signed certificates | Configure internal CA in settings |
| LDAP sync delays | Async job queue | Wait or manually create accounts |
| Service stops after 1-2 weeks | OS temp cleanup | Set `-Djava.io.tmpdir=/path/to/tmpdir` |

### Debug Commands

```bash
# Check API server logs
docker logs -f dependency-track-apiserver

# Verify database connectivity
docker exec -it dependency-track-apiserver \
  curl -s http://localhost:8080/api/version

# Check analyzer status
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/configProperty?groupName=analyzer"

# Monitor mirror status
curl -s -H "X-Api-Key: ${API_KEY}" \
     "${DTRACK_URL}/api/v1/mirror/nvd"

# Health check
curl -s "${DTRACK_URL}/api/version"
```

### Performance Tuning

```yaml
# JVM settings for high-load environments
environment:
  JAVA_OPTIONS: >-
    -Xms8g
    -Xmx16g
    -XX:+UseG1GC
    -XX:MaxGCPauseMillis=200
    -XX:+ParallelRefProcEnabled
```

---

## 10. Reference Files

Additional templates and examples are in the `references/` directory:

**Deployment:**
- `references/deployment/docker-compose-production.yaml` - Docker Compose for production
- `references/deployment/helm-values.yaml` - Helm values for v4.x (Chart v0.40.0)
- `references/deployment/helm-values-hyades.yaml` - Helm values for Hyades v5.x (Chart v0.10.0)
- `references/deployment/kubernetes-manifests/` - Raw Kubernetes manifests

**CI/CD Integration:**
- `references/cicd/jenkinsfile` - Jenkins Pipeline example
- `references/cicd/github-action.yaml` - GitHub Actions workflow
- `references/cicd/gitlab-ci.yaml` - GitLab CI/CD pipeline
- `references/cicd/azure-pipeline.yaml` - Azure DevOps Pipeline

**API & Scripts:**
- `references/api/python-client.py` - Python client library example
- `references/api/bash-scripts/` - Shell scripts for common operations

**Policies:**
- `references/policies/security-policies.json` - Security policy templates
- `references/policies/license-policies.json` - License policy templates
- `references/policies/operational-policies.json` - Operational policy templates

**Troubleshooting:**
- `references/troubleshooting.md` - Common issues and solutions

---

## Related Skills

- `defectdojo-skill` - Vulnerability management integration
- `argocd-skill` - GitOps deployment of Dependency-Track
- `quality-guardian` - Security testing integration

---

## Gotchas

- **BOM upload without `projectVersion` overwrites the previous BOM:** Uploading to the same project name without a distinct version replaces the inventory instead of versioning it. Always pass `projectVersion` (commit SHA or semver) and consider `parentName/parentVersion` for monorepos.
- **OSS Index analyzer runs on a 6-hour schedule, not on upload:** A fresh SBOM may show zero vulnerabilities for hours because the analyzer hasn't swept it yet. Re-upload the BOM or wait — don't conclude "clean" from the immediate UI state.
- **`/api/v1/oidc/mapping` expects UUID strings, not objects:** Sending `{"team": {"uuid": "..."}, "group": {"uuid": "..."}}` returns 400. Use `{"team": "<uuid>", "group": "<uuid>"}`. OIDC groups only appear in the UI after a user from that group logs in once.
- **First-time startup mirrors NVD for 10-30+ minutes:** During this window the API works but vulnerability matching is incomplete. Do not benchmark, do not ship pipeline integration until logs show NVD mirror complete — early scans look falsely clean.
- **API server needs 4.5GB RAM minimum:** Below that the JVM OOMs mid-mirror leaving a half-initialized database. Recovery often requires wiping `/data` and restarting. Set Helm `apiServer.resources.requests.memory` accordingly before scaling.
- **NGINX Ingress default body size rejects large SBOMs as 413:** Monorepo or container BOMs frequently exceed 1MB. Add `nginx.ingress.kubernetes.io/proxy-body-size: "100m"` to the ingress annotations.