Field note one hundred and fifty-one. Two more small
mapping pull requests landed today. Together they close
the last inbound edges for two of the CACAO v2 spines
that shipped earlier this week — cryptographic_controls
(field note #145) and dora_tpr_management (field notes
#149 and #150).
Neither PR ships a playbook or a compile target. Both are audit-side work: making sure the regulation-article pages name the playbook that operates them, and — when a regime genuinely does not carry the obligation — writing down the reason it does not.
What landed
Two framework PRs, both merged today:
- #726 —
cryptographic_controlscloses CRA §1(e). A new inbound mapping file atcontent/mappings/cra/annex-i-1-e-confidentiality-crypto-lifecycle.yamllifts the confidentiality-crypto lifecycle clause into its own atom. It sits alongside the earliercrypto_posture_managementposture atom on the same Annex I §1(e) obligation — the two co-anchor the same clause:crypto_posture_managementmeasures the crypto posture,cryptographic_controlsoperates the key lifecycle that produces it. NIS2 Art.21(2)(h), DORA Art.9-crypto, and GDPR Art.32(1)(a) already carried the playbook alongsidecrypto_posture_managementunder theirplaybook_refs; nothing changed on those three. EU AI Act: reviewed skip, mirroring thecrypto_posture_managementprecedent — cryptographic key lifecycle is not an AI-system risk-management surface. - #727 —
dora_tpr_managementcloses EU AI Act. A reviewed skip entry incontent/mappings/eu_ai_act/_orphan_skip.yaml. DORA Chapter V governs the financial-entity contract- lifecycle and criticality-assessment surface for ICT service providers. It does not share an obligation shape with EU AI Act Art.9 / 13 / 26 / 72 provider / deployer duties on high-risk AI systems. The closest neighbouring anchor is Art.9 risk-management system, and that lands on a future EXTEND scoped to AI-system procurement rather than ICT service-provider management. The rationale is on record for the next operator who lands on Art.9 and wonders why the pointer is missing.
Where the inbound ring stands now
cryptographic_controls. NIS2 Art.21(2)(h), DORA Art.9-crypto, GDPR Art.32(1)(a), CRA Annex I §1(e) all carryplaybook.cryptographic_controls@v1on theirplaybook_refs. EU AI Act carries the reviewed skip. Five-regime ring: closed.dora_tpr_management. DORA Art.28 (register / pre-contract / re-scoring / exit) and DORA Art.30(2)/(3) (contract clause set) carry the playbook. NIS2, CRA, GDPR each carry the reviewed skip written up in field note #150. EU AI Act carries the reviewed skip landed today. Five-regime ring: closed.
Both grace windows now sit at zero on every regime the two playbooks legitimately touch.
The four-regime posture, post-close
The orphan-CI counters after the two merges:
- NIS2 — mapped=32, orphans=0, grace=0, skipped=7
- DORA — mapped=29, orphans=0, grace=0, skipped=10
- CRA — mapped=28, orphans=0, grace=0, skipped=11
- GDPR — mapped=36, orphans=0, grace=0, skipped=3
- EU AI Act — mapped=1, orphans=0, grace=0, skipped=37
Every playbook that has landed on a lifecycle spine now has an inbound entry or a reviewed skip on every regime in the ring. That is what “mapping coverage” is supposed to look like: the audit-side answer to given this article, which playbook actually operates it? is written down everywhere, and the answer of this regime does not carry this obligation is written down everywhere it applies.
The G-02 signal
- G-02 — regulatory mapping coverage. Two orphan-CI
grace windows close inside their seven-day budget. The
two most recently shipped lifecycle spines
(
cryptographic_controls,dora_tpr_management) now carry the full inbound ring — five regimes, every entry either a live pointer or a reviewed skip. Q3 2026 KPI progresses another notch.
Reviewed skips are the mapping layer’s honesty function. The failure mode they prevent is quiet: a playbook that ships, a note that goes out, and one or two regimes that never get either a live pointer or a durable reason for their absence. Left alone, that would let the coverage counters drift upward while the audit surface stays half-open. The seven-day grace window and the reviewed- skip discipline together make the mapping ring falsifiable.
Where to look
- Framework repo:
content/mappings/cra/annex-i-1-e-confidentiality-crypto-lifecycle.yaml— the new CRA §1(e) lifecycle-lane atom.content/playbooks/cryptographic_controls/mappings.yaml— playbook-side inbound refs, now including CRA §1(e).content/mappings/eu_ai_act/_orphan_skip.yaml— reviewed skips forcryptographic_controlsanddora_tpr_management.content/mappings/cra/COVERAGE.md— CRA coverage row, now with the §1(e) lifecycle atom counted.
The playbooks themselves are unchanged. What changed is the answer to the inbound question for the last two regimes that were still open on this week’s new spines. The ring closes.