Craton Shield

Cybersecurity Case

Cybersecurity Case

Project: Craton Shield | Standard: ISO/SAE 21434 Clause 6.4.2 | Version: 0.6.0 | Date: 2026-03-28


1. Purpose

This cybersecurity case aggregates all evidence demonstrating that Craton Shield satisfies cybersecurity requirements for its intended operating context. It serves as the top-level cybersecurity argument per ISO/SAE 21434 Clause 6.4.2.


2. Scope

Item: Craton Shield embedded security runtime platform Boundary: Software component running on a central gateway ECU (NXP S32G3 Cortex-A53 or equivalent). Processes CAN/Ethernet frames via HAL traits, exposes C FFI for integration. Does not control actuators directly. Target context: Series-production passenger vehicles (UN R155 regulated markets).


3. Cybersecurity Claim

Craton Shield provides defense-in-depth protection for automotive embedded systems through:

  • Real-time CAN and Ethernet intrusion detection
  • Policy-driven network firewall
  • Secure OTA update validation (TUF/Uptane)
  • Cryptographic key lifecycle management
  • Measured secure boot chain
  • Runtime integrity monitoring
  • Tamper-evident event logging

All identified threats from the TARA have corresponding mitigations, and all mitigations are verified through automated testing.


4. Evidence Index

4.1 Threat Analysis (TARA)

DocumentReference
Threat Analysis and Risk Assessmentdocs/tara.md

The TARA identifies 26 threat scenarios across six domains (CAN bus, Ethernet, OTA, key management, secure boot, operational) using STRIDE methodology. Risk is assessed using attack feasibility ratings and damage potential per ISO/SAE 21434 Annex E.

4.2 Implemented Mitigations

MitigationCrateTARA Coverage
CAN intrusion detection (rate, ID, sequence, entropy)vs-can-monitor, vs-ids-engine, vs-anomalyT-CAN-01 through T-CAN-05
Ethernet packet inspection (SOME/IP, DoIP, protocol filtering)vs-eth-monitor, vs-ids-engineT-ETH-01 through T-ETH-05
Network firewall (layer 3/4 filtering, rate limiting)vs-netfw, vs-policy-engineT-ETH-02, T-ETH-04
OTA update validation (TUF metadata, rollback protection, signature verification)vs-ota-validatorT-OTA-01 through T-OTA-05
Cryptographic key management (slot-based, zeroize-on-drop)vs-key-manager, vs-cryptoT-KEY-01 through T-KEY-04
Secure boot chain verification (measured boot, PCR extension)vs-secure-bootT-BOOT-01, T-BOOT-02
Runtime integrity monitoring (memory, configuration)vs-integrityT-BOOT-03, T-OPS-01
Tamper-evident audit logging (HMAC-chained)vs-event-loggerT-OPS-02, T-OPS-03

4.3 Security Design Principles

PrincipleImplementation
Memory safety#![forbid(unsafe_code)] on all crates except vs-ffi (audited)
Constant-time operationssubtle crate for comparisons; no secret-dependent branching
Zeroizationzeroize derive on all key material; verified via dedicated tests
Fail-closedAll error paths return deny/reject; no silent degradation
Minimal attack surfaceno_std compatible; no heap allocation in core path
Defense in depthMultiple independent detection layers per threat domain

4.4 Cryptographic Controls

AlgorithmStandardUsage
AES-256-GCMFIPS 197 + SP 800-38DOTA image encryption, secure channel
SHA-256FIPS 180-4Firmware integrity, TUF metadata hashing
HMAC-SHA-256FIPS 198-1Message authentication, key derivation, log chaining
ECDSA P-256FIPS 186-5OTA signature verification, V2X authentication
ECDH P-256SP 800-56Ar3Key agreement for secure sessions

See also: docs/certification/fips-140-3-boundary.md, docs/certification/fips-140-3-kat-plan.md.

4.5 Verification Evidence

CategoryCountReference
Unit tests1,014+cargo test across all workspace crates
Integration tests180+tests/ directory (attack scenarios, crypto vectors, fault injection, concurrency, stress)
Fuzz targets4fuzz/fuzz_targets/ (CAN frame, Ethernet parse, nonce counter, OTA JSON)
Property-based testsPresenttests/property_tests.rs
WCET analysisPresenttests/wcet_stats.rs, benches/wcet_harness.rs
ECU validation suitePresenttests/ecu_validation.rs
Competitive benchmarksPresentbenches/competitive_benchmarks.rs

4.6 CI/CD Controls

ControlToolPurpose
Static analysisclippy (pedantic)Lint enforcement on every commit
Dependency auditcargo auditKnown vulnerability detection (weekly + CI)
License and supply-chain policycargo denyLicense allowlist, advisory DB check, duplicate detection
SBOM generationCI workflowSoftware Bill of Materials for each release
SLSA attestationCI workflowBuild provenance and supply-chain integrity
Branch protectionGitHubRequired reviews, signed commits, status checks
DependabotGitHubAutomated dependency update PRs

4.7 Process and Compliance Documents

DocumentReference
ISO 26262 ASIL-B assessmentdocs/certification/iso-26262-asil-b-assessment.md
ISO/SAE 21434 gap analysisdocs/certification/iso-21434-gap-analysis.md
FIPS 140-3 module boundarydocs/certification/fips-140-3-boundary.md
FIPS 140-3 KAT plandocs/certification/fips-140-3-kat-plan.md
UN R155/R156 evidencedocs/certification/un-r155-r156-evidence.md
Tool qualification reportdocs/certification/tool-qualification-report.md
Penetration test plandocs/certification/penetration-test-plan.md
DFA reportdocs/certification/dfa-report.md
Code review recordsdocs/certification/code-review-records.md
Security policy (SECURITY.md)SECURITY.md
Architecturedocs/architecture.md
Test plandocs/test-plan.md
Requirements traceability matrixdocs/requirements-traceability-matrix.md

5. Residual Risk Assessment

All 26 threat scenarios identified in the TARA have corresponding mitigations implemented and verified. Residual risks are:

  1. Hardware-dependent attacks (side-channel, fault injection on physical hardware) -- mitigated at software level through constant-time operations; full mitigation requires hardware countermeasures outside Craton Shield scope.
  2. Zero-day vulnerabilities in dependencies -- mitigated through automated auditing, SBOM tracking, and 48-hour acknowledgment / 72-hour patch SLA.
  3. Post-quantum threat to ECDSA/ECDH -- mitigated by feature-gated ML-KEM-768 and ML-DSA-65 readiness (experimental, not yet FIPS-approved).

6. Conclusion

This cybersecurity case demonstrates that Craton Shield 0.7.0 satisfies the cybersecurity requirements for its intended deployment context. All TARA-identified threats have implemented and verified mitigations. The evidence chain from threat identification through implementation, verification, and operational monitoring is complete and traceable.