Evaluation

Verification and requirement traceability

Trace reader requirements through architecture, implementation boundaries, and verification evidence.

This is the public requirements-to-evidence view for evaluators and thesis review. It presents reader goals, current component responsibility, and verification without turning the product navigation into a raw SRS or SDD.

Product requirements

IDRequirementImplementation boundaryVerification evidence
FR-01Ingest uploads and supported source evidence without bypassing governanceAPI upload, connector contract, worker ingestion, source ledgerIngestion idempotency, incomplete evidence, connector checkpoint, and adapter suites
FR-02Preserve immutable source revisions and permission evidencePostgreSQL ledger, sealed ACL generations, object storageDatabase immutability, stale-writer, revision conflict, and ACL rotation tests
FR-03Return only content visible to the current actorOIDC binding, OpenFGA, canonical SQL policyIdentity negative tests, OpenFGA model tests, pre-ranking retrieval integration
FR-04Produce grounded Assistant answers with current citationsSecure GraphRAG, evidence-closure recheck, Assistant streamingGraph closure, model mismatch, citation ordering, revoked-citation browser tests
FR-05Govern reusable organizational capability as exact releasesAsset Registry and four profile validatorsDomain integration, review/release/withdrawal, golden POC browser tests
FR-06Deliver Knowledge and Assets to external AI clients safelyMCP protected resource, token exchange, canonical delivery APIIssuer/audience/scope tests, tool/resource schemas, opaque denial, rate limits
FR-07Let administrators govern identities, source mappings, invitations, connections, and auditAdmin API, identity/source ledgers, OpenFGA administrator gateAPI-wide denial, concurrent binding, mapping grant/revoke, credential redaction
FR-08Deploy and recover components independentlyImmutable images, Compose, health/smoke and rollback scriptsCI Gate, image build, deployment smokes, product rollback path

Quality requirements

IDQuality attributeMeasurable interpretationCurrent evidence
QR-01ConfidentialityDenied/cross-tenant evidence cannot affect ranking, graph context, citation, or Asset outputPermission-negative SQL, graph conformance, opaque REST/MCP tests
QR-02IntegrityImmutable revisions, releases, ACLs, package digests, and publication generations cannot be silently rewrittenDatabase guards, exact-pin tests, Skill digest tests, publication CAS
QR-03AccountabilitySecurity and Asset actions identify actor, object, decision, and immutable coordinates without raw secretsAppend-only permission audit and sanitized Asset trace tests
QR-04AvailabilityRetries are bounded; restarts resume; failed projection/deploy preserves a known-good predecessorConnector checkpoint, leased job, atomic publish, health/rollback checks
QR-05UsabilityPrimary reader and operator journeys remain keyboard/mobile accessible and observableWeb and docs browser suites plus automated accessibility smoke
QR-06PerformanceRequests and background work have explicit resource/time/cardinality boundsConfigured limits and deterministic tests; production latency targets not yet measured

Design evidence map

The public software-design view is distributed by reader concern:

Evidence discipline

Automated evidence is named in the repository's mirrored domain coverage files. Runtime availability is verified separately from behavioral tests. Recorded external-provider responses are labelled as such, and missing live or measured evidence remains visible in Limitations.

Last updated on