Document SC 27 Security Evidence Brief
Subject Agentic AI security gaps in ISO/IEC 27090 (FDIS)
Prepared by Kade Cowper — TUARA KURI LIMITED
Date 2026-05-21
Intended for ISO/IEC JTC 1 SC 27 — consideration for next revision cycle or companion Technical Report
Reference agenticrail.nz/spec/receipt-schema.json — formal receipt schema (JSON Schema Draft 2020-12)

Agentic AI Security Gaps in ISO/IEC 27090

ISO/IEC 27090 (FDIS, February 2026) defines the threat taxonomy for AI model security. Agentic AI execution sequences — autonomous systems that commit to ordered multi-step actions with real-world effect — fall outside its scope. Four threat categories are unaddressed. This brief documents each gap and the production mechanism that addresses it.

1. ISO/IEC 27090 Scope

27090's threat taxonomy covers: adversarial evasion, data and model poisoning, model theft and extraction, membership inference, model inversion, prompt injection, and output injection. Scope is deliberate, malicious attacks on AI models and training data. Agentic AI execution sequences are not within scope — the standard predates their production deployment. Four gaps result.

2. Identified Gaps for Agentic AI Systems

Gap 1 — Replay Attacks on Agent Execution Sequences

Gap in ISO/IEC 27090

27090 classifies prompt injection as a single-request threat. It has no threat category for sequence replay — submission of a prior step's valid payload to manipulate an agent's stateful execution context. A fresh nonce does not protect against this; step position in an ordered sequence is the missing invariant. Neither vector appears in 27090's taxonomy.

AgenticRail Mechanism

Per-sequence Durable Object maintains nonce set and step counter. Nonce reuse → REPLAY_NONCE regardless of step position. Step submitted out of counter order → SEQUENCE_VIOLATION. Timestamp window: |ts_ms − server_now| ≤ 300,000 ms enforced independently → STALE_TIMESTAMP. Three orthogonal barriers; all must be defeated simultaneously.

Gap 2 — Pre-Execution Cryptographic Receipt Architecture

Gap in ISO/IEC 27090

27090 Clause 7 references audit logging as a mitigation control with no format, timing, or integrity requirement. A DENY event in an agent gate leaves no forensic record under 27090 — nothing distinguishes "gate ran and refused" from "gate was not called." Pre-execution receipts are absent from the control set.

AgenticRail Mechanism

HMAC-SHA256 signed receipt written to immutable R2 storage before action executes — on every decision including DENY and HALT. pack_id = SHA-256(canonicalJson(decision)). Seven enumerated denial codes carried on DENY receipts. Formal schema: agenticrail.nz/spec/receipt-schema.json (JSON Schema Draft 2020-12, published 2026-05-17).

Gap 3 — Tamper-Evident Sequence Integrity

Gap in ISO/IEC 27090

27090 addresses integrity of training data and model assets. It specifies no chain-linkage requirement for enforcement records across a multi-step sequence. A modified receipt at step N passes single-receipt HMAC verification — only cross-receipt linkage reveals the tampering. Sequence-level integrity is unaddressed.

AgenticRail Mechanism

prev_receipt_id on every receipt = pack_id (SHA-256) of the preceding receipt. Modifying any receipt changes its hash, breaking the reference in every subsequent receipt. Self-verifying; no trusted third party required.

Gap 4 — Permanent Sequence Sealing

Gap in ISO/IEC 27090

27090 specifies no control for permanent closure of an agent execution sequence. A completed sequence without a sealing mechanism is vulnerable to credential-compromise-then-inject: valid credentials obtained post-completion allow retrospective step submission. The threat is absent from the standard.

AgenticRail Mechanism

settle step retires sequence_id in the Durable Object. All subsequent submissions return SEALED_SEQUENCE. No unsealing path exists by design. sealed: true HMAC-signed into the final receipt.

3. Gap Summary

Gap27090 CoverageAgenticRail MechanismDenial Code
Sequence replayNot addressedNonce uniqueness + timestamp freshnessREPLAY_NONCE / STALE_TIMESTAMP
Pre-execution receiptNot addressedHMAC-signed receipt before executionpack_id + signature
Chain integrityNot addressedprev_receipt_id SHA-256 chainHash break on tamper
Sequence sealingNot addressedIrreversible settle stepSEALED_SEQUENCE

4. Production Evidence

Production since February 2026:

MetricValue
Enforcement decisions1,000,000+
Sequence violations missed0
Receipt signing algorithmHMAC-SHA256 (hmac-sha256); Ed25519 implemented, pending key provisioning
Receipt key IDk1_2026-02-22_01
Live verificationreport.agenticrail.nz/report — enter demo-loan-mpamzf82, no login required
Formal receipt schemaagenticrail.nz/spec/receipt-schema.json — JSON Schema Draft 2020-12
DeploymentCloudflare Workers + Durable Objects + R2 + KV
References
27090 addresses model-layer threats. The four gaps above are execution-sequence threats — a different attack surface. Candidate for next-revision extension or companion Technical Report.
ISO/IEC FDIS 27090 — Cybersecurity — Artificial Intelligence — Guidance for addressing security threats and compromises to artificial intelligence systems. FDIS stage, February 2026. iso.org/standard/56581.html

AgenticRail Receipt Schema — slp8_receipt_v2. JSON Schema Draft 2020-12. Published 2026-05-17. agenticrail.nz/spec/receipt-schema.json

AgenticRail Enforcement Specification v1.0 — Deterministic gate, receipt structures, sequence enforcement, cryptographic verification model. Published 2026-05-17. agenticrail.nz/spec/

Live Verificationreport.agenticrail.nz/report — public verification of a complete receipt chain. No login required.

Entity — TUARA KURI LIMITED, 431 Omanaia Road, RD 3, Kaikohe 0473, New Zealand. hello@agenticrail.nz