P179 — AIEP — Evidence Domain Coverage Map Protocol

Publication Date: 2026-03-27 Status: Open Source Prior Art Disclosure Licence: Apache License 2.0 Author/Organisation: Phatfella Ltd Schema: AIEP_OS_SPEC_TEMPLATE v1.0.1 — https://aiep.dev/schemas/aiep-os-spec-template/v1.0.1

Framework Context

[0001] This disclosure operates within an Architected Instruction and Evidence Protocol (AIEP) environment as defined in United Kingdom patent application number GB2519711.2, filed 20 November 2025, the entire contents of which are incorporated herein by reference.

[0002] The present disclosure defines a protocol for computing and publishing a structured DomainCoverageMap — a multi-dimensional representation of the density, freshness, trust tier distribution, and gap profile of evidence artefacts across the AIEP evidence taxonomy (P160) — enabling corpus operators and reasoning chains to identify domains of high evidence confidence, domains of sparse coverage, and domains with significant evidence volatility risk.

Field of the Disclosure

[0003] This disclosure relates to corpus coverage analysis and gap identification protocols for evidence taxonomies in governed artificial intelligence systems.

[0004] More particularly, the disclosure concerns: a DomainCoverageMap schema; a CoverageCell representing a single taxonomy domain leaf node; a CoverageMapGenerator process; a gap surface algorithm; a coverage map publication and versioning mechanism; and the integration of coverage data with evidence package assembly (P174) and quality metrics (P172).

Background

[0005] A reasoning chain operating over an evidence corpus may receive well-formed, high-trust evidence for some domains while drawing on sparse or stale evidence for others. The differential quality across taxonomy domains is not visible to the reasoning chain from individual artefact metadata alone — the chain would need to enumerate artefacts across all relevant taxonomy nodes to form a picture of coverage density.

[0006] A pre-computed DomainCoverageMap provides this aggregated view at corpus level, updated on a regular schedule. Reasoning chains can consult the map before executing queries to calibrate confidence in their evidence base: a query in a domain with a coverage density of 0.1 (very sparse) and a median freshness score of 0.3 (stale) warrants lower reasoning confidence than a query in a domain with density 0.9 and freshness 0.95.

[0007] Coverage maps also guide corpus curation priorities: domains with low density or high volatility are surfaced to corpus operators for targeted ingestion campaigns.

Summary of the Disclosure

[0008] CoverageCell Schema: Each taxonomy leaf node in the evidence taxonomy (P160) is represented by a CoverageCell:

• taxonomy_node_id — the leaf node identifier in the evidence taxonomy (P160)
• artefact_count — total active artefacts in this taxonomy node
• artefact_count_by_trust_tier — breakdown by trust tier (P124): TIER_1, TIER_2, TIER_3, TIER_4, UNSCORED
• median_freshness_score — median freshness score (P147) across all active artefacts in this node
• minimum_freshness_score — minimum freshness score (the stalest artefact in the node)
• artefact_count_with_governance_warning — count of artefacts in this node carrying an active GOVERNANCE_WARNING annotation (P176)
• volatility_class_distribution — distribution across volatility classes (P163): STABLE, SLOW_DECAY, MODERATE_DECAY, RAPID_DECAY, EPHEMERAL
• coverage_density — a computed score in [0.0, 1.0] expressing the ratio of actual artefacts to the reference target artefact count for this taxonomy domain (see [0009])
• gap_severity — NONE (density ≥ 0.8), MINOR (0.5–0.79), MODERATE (0.2–0.49), CRITICAL (< 0.2)
• freshness_risk — LOW (median freshness ≥ 0.7), MODERATE (0.4–0.69), HIGH (< 0.4)
• computed_at — ISO 8601 timestamp of the most recent CoverageCell computation

[0009] Coverage Density Computation: Each taxonomy leaf node in the taxonomy registry (P160) carries an optional reference_artefact_target — the minimum artefact count that the corpus operator designates as representing adequate coverage for this domain. Where reference_artefact_target is not set, a network-wide default (configurable; default: 10 artefacts per leaf node) is used. coverage_density = min(1.0, artefact_count / reference_artefact_target).

[0010] DomainCoverageMap Schema: The aggregate map:

• map_id — SHA-256 of canonical serialisation of all CoverageCells
• generated_by — node fingerprint (P46)
• generated_at — ISO 8601 timestamp
• taxonomy_version — version of the evidence taxonomy (P160) against which the map was computed
• cells — list of CoverageCell, one per active taxonomy leaf node
• global_gap_count — count of taxonomy leaf nodes with gap_severity: CRITICAL
• global_coverage_density — mean coverage_density across all leaf nodes
• map_signature — cryptographic signature by the generating node

[0011] Gap Surface Algorithm: The CoverageMapGenerator runs the gap surface algorithm to identify and rank CRITICAL and MODERATE gap cells:

1. Sort CoverageCells by gap_severity (descending) then by freshness_risk (descending)
2. For each CRITICAL cell, generate a CoverageGapAlert record identifying the taxonomy domain, the deficit artefact count, and the freshness risk
3. Publish CoverageGapAlerts to the Evidence Quality Metrics system (P172) and to the governance node (P89) for corpus curation prioritisation

[0012] CoverageMap Publication: The DomainCoverageMap is computed by the CoverageMapGenerator on a configurable schedule (default: daily). Generated maps are published to: (a) the corpus operator dashboard; (b) all registered mirror nodes (P60–P63); and (c) the Evidence Quality Metrics report endpoint (P172). Reasoning chains may query the current map via GET /evidence/coverage-map on the evidence API, receiving the most recently generated map.

[0013] Coverage Map in Evidence Package Assembly: The EvidencePackage AssemblyRequest (P174) optionally includes include_coverage_context: true, causing the PackageAssembler to include a summary of CoverageCells relevant to the corpus_scope of the package. This allows the receiving reasoning chain to contextualise its evidence base without a separate map query.

ASCII Architecture

Evidence Corpus
(artefact_count, trust_tiers, freshness, volatility per taxonomy node)
        │
        ▼
┌───────────────────────────────┐
│  CoverageMapGenerator         │
│  - counts artefacts per node  │
│  - computes density, risk     │
│  - runs gap surface algorithm │
└──────────────┬────────────────┘
               │ DomainCoverageMap
       ┌───────┼────────────────────────┐
       │       │                        │
       ▼       ▼                        ▼
  Operator  Mirrors              Evidence API
  Dashboard (P60-P63)            GET /evidence/coverage-map
                │
                ▼
       CoverageGapAlerts ──▶ Quality Metrics (P172)
                         ──▶ Governance Node (P89)
                             (curation prioritisation)

Operational Detail

[0014] Taxonomy Hierarchy Rollup: In addition to leaf-node CoverageCells, the DomainCoverageMap includes aggregated intermediate-node cells: for each non-leaf taxonomy node, the cell aggregates the statistics of all descendant leaf nodes. Intermediate-node cells are annotated with is_aggregated: true to distinguish them from leaf cells.

[0015] Historical Coverage Tracking: Previous DomainCoverageMaps are retained for a configurable period (default: 90 days). The coverage API supports GET /evidence/coverage-map/history?from={date}&to={date} returning the sequence of maps in the requested window, enabling trend analysis: operators can observe whether critical gaps are being closed by ingestion activity.

[0016] Multi-Node Coverage Federation: Where multiple nodes hold corpus partitions (e.g. by taxonomy domain shard), a federated CoverageMapGenerator collects CoverageCell data from each node and merges cells for shared taxonomy domains, producing a unified DomainCoverageMap over the full corpus. Merge conflicts (e.g. differing computed artefact_count for the same taxonomy node) are flagged in the merged map as MERGE_DISCREPANCY cells requiring investigation.

Claims-Exclusion Notice

This specification is published as open-source prior art. No patent claims are asserted by the author in respect of the mechanisms described. Any third party seeking to patent mechanisms substantially equivalent to those described herein is placed on notice of this prior art disclosure.