Skip to content

MEM SPARQL Consistency Audit

Generated artifact. Do not edit manually.

Generated by: scripts/build_compendia.py Regenerate with: python3 scripts/build_compendia.py --compendium qa-ci-lanes --web-and-tex Source of truth: docs/qa-ci/lanes-source/

Language / Sprache: English (EN) · Deutsch (DE) | PDF: English PDF · Deutsches PDF

Compendium: §1 CI Architecture · §2 Quality Dashboard · §3 Mapping Workbench · §4 Package Readiness · §5 Release-Model Conformance · §6 Dual Release Contracts · §7 Relation Checks · §8 Atomic Review · §9 Requires Review · §10 Human Review Gates · §11 MEM SPARQL Audit · §12 MEM SPARQL Runbook · §13 Source Rationales Runbook · §14 Abi 2026 DeepLink QA

11. MEM SPARQL Consistency Audit

This deliberately non-blocking QA lane compares live curriculum data from the external MEM/FWU ontology endpoint (https://sparql.mem.edufeed.org/sparql) against SkillPilot's retained source-extraction evidence. Because external endpoint coverage across German states remains incomplete, discrepancies act as structured triage signals rather than CI failures; MEM never replaces official source references or CQR-*, atomicity, memory-card, and composition-view validators.

11.1. Sources of Truth, Generated Artifacts, and Execution

  • Config & ledger (source of truth): curricula/DE/Gymnasium/quality/mem-sparql-consistency/canonical-math-poc.config.json and canonical-math-poc.review.jsonl (generator: app/scripts/generateMemSparqlConsistencyAudit.ts).
  • Generated snapshots (never hand-edited): mem-sparql-consistency-audit.md, mem-sparql-consistency-audit.json, mem-sparql-consistency-review-issues.md, and mem-sparql-consistency-review-issues.json.
cd app
npm run quality:mem-sparql-consistency
npm run quality:mem-sparql-consistency -- \
  --endpoint=https://sparql.mem.edufeed.org/sparql
MEM_CFG=curricula/DE/Gymnasium/quality/mem-sparql-consistency
npm run quality:mem-sparql-consistency -- \
  --config="$MEM_CFG/canonical-math-poc.config.json"

The command exits 0 even when open triage items exist, failing only on local script or configuration errors (initial PoC record: MEM SPARQL PoC, 2026-06-01; operational checklist: 12).

11.2. Fingerprint-Bound Review Ledger and Expansion Criteria

Every JSON review-queue entry carries an itemFingerprint computed from current MEM and local evidence:

{
  "schemaVersion": 1,
  "auditId": "canonical-math-poc",
  "itemId": "triage-a3e85d9ef77a",
  "itemFingerprint": "sha256:...",
  "status": "mem_feedback_candidate",
  "reviewedAt": "2026-06-01",
  "reviewer": "initials-or-tool",
  "reason": "Short evidence-based decision.",
  "followUpRef": "optional issue, PR, or MEM feedback reference"
}
  • Supported status values: not_an_issue, local_extraction_fix_needed, mem_feedback_candidate, source_version_gap, deferred.
  • Per-run ledger validation: Matching itemId + itemFingerprint annotates the entry as decided (except deferred, which stays open); drifted fingerprints surface as stale_review; unknown, duplicate, or malformed records appear under reviewLedger.diagnostics.
  • Expansion criteria: Add a scope only when MEM exposes concrete plan data (beyond vocabulary), graph and class IRIs are known, SkillPilot holds a retained source-extraction artifact for the same scope, and stable issue IDs can be triaged without blocking CQR-* readiness rules.