Completeness means having enough of the right information to support an intended decision. It depends on the decision and affected population—not the volume of data, documents, or tokens available. section 5.6
Assess whether the decision has the information it requires:
- Evidence, constraints, exceptions, and provenance. section 5.6
- Confidence, policy state, consent, and review history. section 5.6
- Relevant populations, geographies, languages, channels, and operating conditions. section 5.6
Breach analysis
Late-delivery records alone may be insufficient: the analysis may also need the amendment, agreed-volume condition, change-control provision, supplier correspondence, contractual exceptions, causation evidence, and evidence of loss. section 5.6
Table 5.6 Verifying and managing agentic data quality
|
Quality dimension
|
What you should verify and manage
|
|
Accuracy
|
Reconcile source data with authoritative records, test
transformation fidelity, and run versioned evaluation datasets against
retrieval, grounding, factual claims, tool use, and expected outcomes.
|
|
Completeness
|
Test whether required fields, sources, evidence,
exceptions, and policy inputs are present for the decision surface.
|
|
Timeliness
|
Monitor freshness, effective dates, supersession, pipeline
delay, retrieval-index lag, and context expiry.
|
|
Consistency
|
Detect conflicting values, duplicates, incompatible
formats, broken relationships, and disagreement across sources.
|
|
Traceability
|
Connect sources, transformations, retrieval, model and
tool activity, policy decisions, and the final decision episode.
|
|
Privacy and confidentiality
|
Discover, classify, mask, redact, and restrict personal,
privileged, confidential, or regulated information.
|
|
Compliance
|
Evaluate policy conditions, processing restrictions,
evidence requirements, approval routes, and control outcomes.
|
Use the completeness row to check whether required fields, sources, evidence, exceptions, and policy inputs are present for a decision. section 5.6






