TISAX VDA-ISA Catalogue

Stefan Coetzeecreated 2026-05-12updated 2026-06-086 min readtisax-catalogue
Vault note, not reviewed against the source. Written in the knowledge vault on 2026-05-12 by models working with Stefan Coetzee and published as it stands, with private addresses, e-mail addresses and an employer name redacted. Check claims against the primary source before relying on them.

The control set against which TISAX assessments are conducted. Authored by VDA, current major version VDA-ISA 6.0 (late 2023) with maintenance updates. Three modules, maturity-level scoring per control, OEM-specified protection requirements per label.

Catalogue structure

The VDA-ISA workbook is an Excel-format spreadsheet containing three modules:

  • Information Security — primary module, required for most TISAX labels.
  • Prototype Protection — automotive-specific, required when handling pre-series components or design data.
  • Data Protection — GDPR-aligned, required when processing personal data on behalf of OEM.

Each control row in the catalogue contains:

  • Control number and title
  • Control objective
  • Reference to applicable ISO 27001 Annex A control (where mapping exists)
  • Maturity-level criteria (per level 0-5)
  • Notes and clarifications
  • Self-assessment scoring fields
  • Auditor scoring fields
  • Comments and evidence reference fields

Maturity levels (SPICE-style 0-5)

Each control is scored across the same six maturity levels:

  • 0 — Incomplete — control not implemented or partially implemented in an ad-hoc way that does not meet the objective.
  • 1 — Performed — control implemented and meets the objective, but informally; depends on individual effort. No process discipline.
  • 2 — Managed — control planned, tracked, verified. Process exists. Outcomes consistent.
  • 3 — Established — control documented as part of standard organizational process. Tailored from a standard. Consistent across the org.
  • 4 — Predictable — control quantitatively managed. Performance is measured, statistically controlled, and predictable.
  • 5 — Optimizing — control continuously improved based on quantitative feedback.

Target maturity is typically 3 for AL2 and AL3 assessments. The OEM-specified protection requirement determines whether a higher target applies to specific controls.

Maturity-level criteria are defined per control; reaching level 3 on control X may require different evidence than level 3 on control Y. The workbook provides the per-control criteria.

Information Security module

The largest module. Aligned with ISO/IEC 27001:2022 Annex A four-theme layout in VDA-ISA 6.0:

  • Organizational controls — policies, roles, asset management, classification, supplier relationships, incident management, regulatory compliance.
  • People controls — screening, awareness, employment terms, post-employment obligations.
  • Physical controls — perimeter, entry, environmental, equipment, off-premises, disposal.
  • Technological controls — endpoints, access, malware, vulnerability, configuration, deletion, masking, DLP, backup, logging, monitoring, network, cryptography, SDLC, secure coding, change management.

Control count varies slightly between VDA-ISA versions but is in the 40-50 range overall — narrower than ISO 27001's 93 Annex A controls. The catalogue selects controls considered automotive-supplier-relevant and folds related controls together.

Headline controls (illustrative, not the full list):

  • Information security policy
  • Information security responsibilities
  • Asset inventory (including information assets)
  • Information classification scheme
  • Access control management (incl. privileged access)
  • Cryptographic controls
  • Physical access controls
  • Employee onboarding and offboarding
  • Training and awareness
  • Network segmentation
  • Logging and monitoring
  • Vulnerability and patch management
  • Backup and recovery
  • Change management
  • Supplier security management
  • Incident response
  • Business continuity
  • Cloud service use (VDA-ISA 6.0 expansion)
  • Configuration management (VDA-ISA 6.0 explicit)
  • Secure software development
  • Outsourced development oversight

VDA-ISA 6.0 introduced or strengthened controls around:

  • Cloud services and shared-responsibility models
  • Configuration baselines and drift detection
  • Secure coding and SDLC integration
  • Information deletion (including vendor-side commitments)
  • Monitoring and anomaly detection
  • AI / new-technology consideration (light-touch reference, not deep)

Prototype Protection module

Automotive-specific. Covers the handling of:

  • Pre-series components — physical parts before production launch
  • Prototype vehicles — test mules, validation vehicles, pre-production builds
  • Design data — CAD, simulation, test data, supplier-provided component specifications
  • Test environments — dynos, environmental chambers, proving grounds
  • Photographic and video evidence — restrictions on capturing and storing imagery

Typical Prototype Protection controls:

  • Restricted-access rooms — secured spaces with badge / biometric entry, log retention
  • No-photo / no-mobile zones — physical signage, device check-in, occasional searches
  • Camouflage requirements — for vehicle prototypes leaving controlled premises
  • Vendor / visitor escort policies — non-employees escorted at all times in prototype areas
  • Air-gapped networks — for CAD and simulation data on prototype projects
  • Time-bounded retention — pre-series data deleted on prescribed schedule post-launch
  • Travel and transport rules — prototype components in transit with security measures
  • Subcontractor / sub-supplier flow-down — Prototype Protection requirements pass through

OEM contracts often specify which Prototype Protection sub-label is required, calibrated to the sensitivity of the supplier's prototype access.

Data Protection module

GDPR alignment. Required when supplier processes personal data on behalf of OEM (Article 28 processor scenarios). Controls cover:

  • Lawful basis for processing
  • Purpose limitation
  • Data minimization
  • Accuracy of personal data
  • Storage limitation (retention schedules)
  • Integrity and confidentiality (technical and organizational measures, TOM)
  • Accountability (records of processing activities)
  • Data subject rights (access, rectification, erasure, portability, objection)
  • Breach notification processes
  • Data protection impact assessments for high-risk processing
  • Cross-border transfer mechanisms (SCCs, adequacy decisions, additional safeguards)
  • Sub-processor management (controller approval, flow-down obligations)

Data Protection has two label tiers:

  • Standard — for general personal-data processing
  • Special-category data — when processing data subject to GDPR Article 9 (health, biometric, ethnic origin, political opinion, etc.)

Scoring and audit conduct

For each applicable control:

  1. Supplier self-scores using the workbook. Selects target level (per OEM requirement) and self-assesses current level. Documents evidence per level achieved.
  2. Audit provider scores independently in AL2 / AL3 assessments. Reviews evidence, conducts interviews, observes practice.
  3. Reconciliation — significant gaps between self-assessment and auditor scoring become discussion points. Often the auditor moderates the self-assessment downward; significant downward moderation is a yellow flag.
  4. Major findings — controls where the audited maturity is below the target maturity for the label. Must be remediated for label issuance; remediation typically allowed within an agreed window.
  5. Minor findings — controls slightly below target but with clear remediation path. May be issued with conditional label.
  6. Observations — items not blocking label issuance but flagged for attention at next assessment cycle.

Common assessment patterns

  • AL2 + Info Sec High — most common combination. Typical maturity targets level 3.
  • AL3 + Info Sec High + Prototype Protection — for development-partner-tier suppliers. AL3 brings on-site audit.
  • AL3 + Info Sec Very High + Prototype Protection + Data Protection — for strategic-partner-tier with sensitive prototype access and PII processing.
  • Multi-site assessments — large suppliers cover multiple sites in one assessment cycle; ENX portal records site-level labels.

VDA-ISA versioning and transitions

Catalogue revisions affect existing labels in a calibrated way:

  • New assessments after the revision use the current catalogue.
  • Existing labels remain valid until their original three-year expiry.
  • Renewal assessments at expiry use the current catalogue.
  • Major revisions (4.x → 5.0, 5.x → 6.0) sometimes trigger ENX-announced transition windows for already-mid-cycle suppliers.

SRE and AI-agent fit notes

  • Cloud-service controls (VDA-ISA 6.0) — model API providers (Anthropic, OpenAI, Google) are cloud services for catalogue purposes. Configure DPAs, no-training-data-use settings, enterprise tiers.
  • Configuration management — extends naturally to AI agent system prompts, tool scopes, model versions. Treat as configuration artefacts under VDA-ISA 6.0 expectations.
  • Secure coding and outsourced development — apply to AI-coding-assistant output. The catalogue does not yet name this explicitly; mature implementations cover it within secure-development controls.
  • Information deletion — automotive-specific retention controls plus VDA-ISA general deletion controls combine to require vendor-side retention proof. Model providers' zero-retention enterprise tiers are the path.
  • Prototype Protection × AI tools — pre-series data must not flow to AI tools that retain or use it for training. Default ChatGPT, Claude.ai, Copilot configurations typically fail Prototype Protection criteria. Enterprise configurations with retention controls + DPAs are the workable path.
  • Monitoring and anomaly detection — VDA-ISA 6.0 expansion lands close to needed AI-agent monitoring practice. Use the control to anchor agent action logging and anomaly detection.

Stefan-context implementation sketch

Solo / small-team operation against automotive client:

  • Use VDA-ISA workbook as self-assessment tool even without formal assessment. Internal target maturity level 3 against the Info Sec module covers most engagement needs.
  • For engagements touching prototype data: apply Prototype Protection controls regardless of formal label requirement. Air-gapped (or strongly segregated) work environment, no-AI-tool-retention configuration, time-bounded retention.
  • For engagements touching customer PII: apply Data Protection module controls. GDPR-aligned processing records, sub-processor (vendor) documentation, retention schedules.
  • Map self-implementation back to ISO 27001 cluster atoms where controls overlap; avoid duplicating documentation.

See also