Skip to content
Agent packs

Lending · Agent pack

Credit application intake

Cadre pack · Onramp

Assembles a verified underwriting package or returns an incomplete file for resolution.

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

Checks supplied application and credit-file data, returning an underwriting package or explicit data-quality and incompleteness flags.

Case arrives with
Applicant · Product · Requested amount · Term
Allowed outcomes
Complete · Incomplete
Where it fits
Application to Decision
Decision logic
  1. 01Verifies a submitted application and assembles the package the underwriter works from: required fields, the bureau pull, the pre-loan debt-to-income ratio.
  2. 02Makes no credit decision and is designed not to receive or seek protected characteristics; its role is limited to the supplied credit-relevant facts.
  3. 03Surfaces data-quality problems as flags: missing fields, thin files, and a disputed derogatory are routed for resolution instead of being silently treated as settled inputs.
  4. 04Marks the package not ready for underwriting when it is incomplete, unscorable, or carries an unresolved dispute that bears on the decision.
Must not
Use protected-basis fields in the decision path.
Human owner
Data conflicts and incomplete applications
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
Applicant
Dana Whitfield
Product
auto
Requested amount
$31,000
Term
60 months
Credit score
768 (prime)
Purpose
vehicle purchase
View the case data

Case key L1-clean-auto

{
  "applicationId": "L1"
}

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.

  • prime auto, complete, scored

    ready

    Case key

    L1-clean-auto
  • prime card, complete

    ready

    Case key

    L7-clean-card
  • no score / thin file

    flag, not ready

    Case key

    L6-thin-file
  • active FCRA dispute on a derogatory

    flag, not ready

    Case key

    L13-fcra-dispute

Operating fit

Workflow placement and validation

Operating pattern

Decide, refer, or stop

Give the pack a short list of permitted decisions and require referral when the evidence cannot support one.

Decision sequence and workflow placement
  1. 01Receive the case
  2. 02Gather named evidence
  3. 03Apply the decision rules
  4. 04Decide or refer
  5. 05Record the decision and owner
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