This file lists the decisions Azoth took for this system over the period: 5 in total — 1 refused or held before anything reached an AI provider, and 4 let through. Each decision is chained to the previous one by a cryptographic fingerprint, so the integrity verdict above shows whether any entry was changed, added or removed after the fact.
Chain-integrity verdict
INTACT — the sealed decision chain verifies end to end.
Algorithm
sha256(prev_hash || core_canonical)
Refused or held calls
Calls Azoth stopped before they reached a provider, with the rule or reason recorded at the time. Nothing in this list was sent to the AI provider.
#
When (UTC)
Decision
Provider · model
Rule / reason
2
2026-11-06T09:05:00.000Z
blocked_by_policy
—
—
Volume
Sealed decisions
5
Sequence range
1 … 5
Cost (EUR)
0.0125
Energy (Wh)
0.00
Carbon (g CO₂e)
0.00
Decisions by surface
Verdict
Count
governance
2
inference
2
guard
1
Guard-reviewed decisions
Guard decisions in period
1
Verdict
Count
guard_require_approval
1
Obligations — technical evidence coverage
Each row lists an obligation the sealed decisions above are technical evidence for. This is not a determination of compliance.
Framework
Article
Evidence (decisions)
Risk categories
None in this period.
Disclaimer. Azoth produces technical evidence supporting your obligations. It does not certify, guarantee, or attest compliance, and is not legal advice.