Drift Detection

Skill: Drift Detection

> Expertise: Terraform plan-based drift detection, CI scheduling, cloud audit log correlation, drift classification.

When to load

When setting up scheduled drift detection, investigating detected drift, or correlating drift with cloud audit events.

Scheduled Drift Detection (GitHub Actions)

# .github/workflows/drift-detection.yml
name: Drift Detection

on:
  schedule:
    - cron: '0 */6 * * *'   # every 6 hours
  workflow_dispatch:          # manual trigger

jobs:
  detect-drift:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        component: [network, k8s-cluster, databases, iam-roles]
    steps:
      - uses: actions/checkout@v4

      - name: Configure AWS credentials (OIDC)
        uses: aws-actions/configure-aws-credentials@v4
        with:
          role-to-assume: ${{ vars.DRIFT_DETECTOR_ROLE_ARN }}
          aws-region: eu-west-1

      - name: Terraform init
        working-directory: terraform/environments/production/${{ matrix.component }}
        run: terraform init -backend-config=backend.hcl

      - name: Terraform plan (drift check)
        id: plan
        working-directory: terraform/environments/production/${{ matrix.component }}
        run: |
          terraform plan \
            -var-file=terraform.tfvars \
            -detailed-exitcode \
            -out=drift-check.plan \
            2>&1 | tee drift-output.txt
          echo "exit_code=$?" >> $GITHUB_OUTPUT
        continue-on-error: true    # exit 2 = changes, don't fail job

      - name: Classify drift
        if: steps.plan.outputs.exit_code == '2'
        run: |
          # Check for security-sensitive resource changes
          if grep -E "aws_iam|aws_security_group|aws_kms|encryption" drift-output.txt; then
            echo "SEVERITY=INVESTIGATE" >> $GITHUB_ENV
          else
            echo "SEVERITY=REMEDIATE" >> $GITHUB_ENV
          fi

      - name: Create GitHub Issue on drift
        if: steps.plan.outputs.exit_code == '2'
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            const plan = fs.readFileSync('terraform/environments/production/${{ matrix.component }}/drift-output.txt', 'utf8');
            await github.rest.issues.create({
              owner: context.repo.owner,
              repo: context.repo.repo,
              title: `[DRIFT] ${{ matrix.component }} — ${process.env.SEVERITY}`,
              body: `## Drift detected in \`${{ matrix.component }}\`\n\n**Severity:** ${process.env.SEVERITY}\n\n\`\`\`\n${plan.slice(0, 3000)}\n\`\`\``,
              labels: ['infrastructure', 'drift', process.env.SEVERITY.toLowerCase()]
            });

      - name: Alert on INVESTIGATE drift
        if: env.SEVERITY == 'INVESTIGATE'
        uses: slackapi/slack-github-action@v1
        with:
          payload: |
            {"text": "🚨 *INVESTIGATE drift* in `${{ matrix.component }}` — may indicate unauthorized change. "}
        env:
          SLACK_WEBHOOK_URL: ${{ secrets.SLACK_ONCALL_WEBHOOK }}

GitLab CI Scheduled Drift

# .gitlab-ci.yml — drift detection job (triggered by schedule)
drift-detect:
  stage: audit
  rules:
    - if: $CI_PIPELINE_SOURCE == "schedule"
  script:
    - cd terraform/environments/production/${COMPONENT}
    - terraform init -backend-config=backend.hcl
    - terraform plan -var-file=terraform.tfvars -detailed-exitcode
      -out=drift.plan 2>&1 | tee drift-output.txt || EXITCODE=$?
    - |
      if [ "${EXITCODE}" == "2" ]; then
        echo "DRIFT DETECTED in ${COMPONENT}"
        # Post to Slack, create MR, or alert via API
        curl -X POST "$SLACK_WEBHOOK" \
          -H 'Content-type: application/json' \
          --data "{\"text\":\"Drift in ${COMPONENT}\"}"
        exit 1
      fi
  parallel:
    matrix:
      - COMPONENT: [network, k8s-cluster, databases]

Drift Classification Logic

# Parse terraform plan output for drift classification
classify_drift() {
  local plan_output="$1"

  # INVESTIGATE: security-sensitive resources changed
  if echo "$plan_output" | grep -qE \
    "aws_iam_policy|aws_security_group|kms_key|aws_s3_bucket_server_side_encryption|encryption_at_rest"; then
    echo "INVESTIGATE"
    return
  fi

  # INVESTIGATE: resources deleted unexpectedly
  if echo "$plan_output" | grep -q "# .* will be destroyed"; then
    echo "INVESTIGATE"
    return
  fi

  # REMEDIATE: configuration drift (non-security)
  echo "REMEDIATE"
}

Correlating Drift with Cloud Audit Logs

# AWS CloudTrail: who changed the drifted resource?
aws cloudtrail lookup-events \
  --lookup-attributes AttributeKey=ResourceName,AttributeValue= \
  --start-time $(date -d '-24 hours' --iso-8601) \
  --query 'Events[*].{User: Username, Time: EventTime, Event: EventName}' \
  --output table

# GCP Audit Logs
gcloud logging read \
  'resource.type="gce_instance" AND protoPayload.methodName:"compute.instances"' \
  --freshness=24h \
  --format='table(timestamp, protoPayload.authenticationInfo.principalEmail, protoPayload.methodName)'

Drift Suppression (accepted exceptions)

# lifecycle ignore_changes for intentionally managed-outside-TF fields
resource "aws_autoscaling_group" "workers" {
  # ...
  lifecycle {
    ignore_changes = [
      desired_capacity,    # managed by cluster autoscaler, not TF
      min_size,
    ]
  }
}

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: sawrus
• Source: sawrus/agent-guides
• License: MIT
• Homepage: https://sawrus.github.io/agent-guides

Install and usage instructions live in the source repository linked above.