Skip to content
Cadre
Agent packs

Lending · Agent pack

Fair-lending review

Cadre pack · Equipoise

Returns findings on prohibited-basis influence and unsupported inconsistency for accountable review.

Demonstrated on representative cases. Source coverage, thresholds, policy interpretation, escalation design, and performance require institution-specific validation.

Operating dossier

The job, permitted decisions, and owner

Mandate

Produces a bounded review of a supplied credit decision for disparate-treatment and prohibited-basis proxy risk, for accountable fair-lending review.

Case arrives with
Application · Decision · Application · Decision · Product · Decision · Decision
Allowed outcomes
Clear · Concern · Violation
Where it fits
Application to Decision
Decision logic
  1. 01Produces a second-line review of a supplied underwriting decision for disparate-treatment and prohibited-basis proxy risk.
  2. 02The one agent permitted to see monitoring attributes, and only as a lens for outcome testing — never as an input to anyone's credit merit.
  3. 03Tests whether similarly-situated applicants received consistent outcomes; a materially equivalent profile with a better result across a prohibited-basis difference is potential disparate treatment.
  4. 04Also audits the decision itself: a within-policy file declined with no hard basis is a red flag regardless of comparators.
  5. 05Returns an internal clear, concern, or violation finding, names the comparators behind it, and leaves the legal and compliance conclusion to accountable reviewers.
Must not
Invent missing evidence or act outside the permitted outcomes.
Human owner
Findings requiring compliance judgment
Evidence retained
Structured outcome, observed path, and the source evidence used.

Case and record

One case, from supplied evidence to recorded decision

Supplied case

Representative case dataSynthetic or simulated data
Application
L11
Decision · Application
L11
Decision · Product
auto
Decision · Decision
decline
Decision · Score band
near prime
Decision · Dti pct
13
Decision · Ltv pct
102
Decision · Policy basis
score 666 in the near-prime band
Decision · Factors
credit score in the near-prime range, moderate revolving utilization
Decision · Confidence
0.62
View the case data

Case key L11-disparate-treatment

{
  "applicationId": "L11",
  "decision": {
    "applicationId": "L11",
    "product": "auto",
    "decision": "decline",
    "scoreBand": "near-prime",
    "dtiPct": 13,
    "ltvPct": 102,
    "policyBasis": [
      "score 666 in the near-prime band"
    ],
    "factors": [
      "credit score in the near-prime range",
      "moderate revolving utilization"
    ],
    "confidence": 0.62
  }
}

Recorded path and outcome

Cadre reference capture

One frozen, sanitized path through representative data. It is an inspection aid, not a performance claim.

Observed path

  1. Get application

  2. Get policy

  3. Get monitoring

  4. Find similar

  5. Get credit report

The full record continues on the capture page.

Recorded outcome

Violation

Verdict

Open decision record

Evaluation and limits

Evaluation coverage and stopping points

Published evaluation map

Expected decisions, boundaries, and costly failure modes.

These cases show what the pack is asked to decide, when it should stop, and which plausible errors the evaluation is meant to expose. They are not a reliability score or independent validation.

Expected and borderline behavior

Routine judgments, close calls, and named stopping points.

  • within-policy near-prime file declined; similarly-situated L12 (different basis) was approved

    concern/violation

    Case key

    L11-disparate-treatment
  • clean prime approve, no disparity

    clear

    Case key

    L1-clean-approve
  • decline well-supported by hard policy fails, no comparator

    clear

    Case key

    L2-supported-decline

Operating fit

Workflow placement and validation

Operating pattern

Independent review

Separate the work from the party that tests it, then require findings to point back to evidence and a named standard.

Decision sequence and workflow placement
  1. 01Receive completed work
  2. 02Reconstruct the standard
  3. 03Run independent checks
  4. 04State findings and limits
  5. 05Assign accountable review

Wider workflows

Application to Decision·Fair-lending review
Pattern guide

Institution-specific validation

These questions remain open until the pack is fitted to an institution.

Validation questions
  • Policy and legal interpretation
  • Source coverage, quality, and freshness
  • Thresholds and exception calibration
  • Integration and degraded-mode behavior
  • Human approval and escalation design
  • Performance on the institution’s own case mix
Inputs, actions, and outcomes
Give it
  • Application
  • Decision · Application
  • Decision · Product
  • Decision · Decision
  • Decision · Score band
  • Decision · Dti pct
  • Decision · Ltv pct
  • Decision · Policy basis
  • Decision · Factors
  • Decision · Confidence
It does
  • Get applicationget_application
  • Get credit reportget_credit_report
  • Get policyget_policy
  • Get monitoringget_monitoring
  • Find similarfind_similar
  • Record reviewcommitrecord_review
It returns · Verdict
  • Clear
  • Concern
  • Violation
Related packs

Related packs

Credit decision model review

Also in Application to Decision · Backstop

Adverse-action reasons

Also in Application to Decision · Grounds

Consumer credit decisioning

Also in Application to Decision · Spark

Consumer lending disclosures

Also in Application to Decision · Plainspeak