BotCurrent↗
← Back to examples

Official recipe · Tool workflows

Check borrower spreads for covenant exceptions

Source-backed example. Practical decision guidance is proposed and untested; this is not a full implementation guide.

The problem

Underwriting reviewers need to locate policy exceptions and unsupported assumptions without losing the spreadsheet cells and financial evidence behind them.

Agent job

Retrieve borrower financials and peer spreads, inspect the underwriting workbook and prepare a cell-referenced exception brief.

Expected output

A source-linked brief for reviewed underwriting-model and memo updates.

Where this framing comes from
  • Core problem: Editorial purpose framing
  • Agent job: Editorial purpose framing
  • Expected output: Editorial purpose framing
  • Relevant job roles: Editorial purpose framing
  • Benefit type: Editorial purpose framing
Editorial framing and outcome limits

Describes the task’s purpose and intended output; measured benefit, safe unsupervised operation and successful delivery are not established by this framing.

Editorial synthesis based on the linked source. This framing does not add evidence of successful delivery or results.

Case-specific decision guidance · untested

Case-specific decision guidance and first checks are original proposals. The linked source supports the attributed task; this added guidance has not been executed.

Read the attributed source evidence and original caveats below ↓

Why use this

  1. An underwriting reviewer may need to trace a possible covenant exception through financial statements, assumptions and spreadsheet formulas. A narrative summary alone can obscure which input or cell requires review.

  2. This job could prepare a cell-referenced evidence brief that makes unsupported assumptions and policy questions visible. It supports qualified review while leaving model changes, credit judgments and any borrower consequence with authorized humans.

Why an agent / simpler alternative

Useful judgment

Map workbook calculations to supplied policy definitions and source evidence, distinguishing exceptions from missing information.

Simpler alternative

Use established spreadsheet controls when calculations and policy tests are already standardized.

Unsuitable when

The workflow would decide credit eligibility, infer sensitive borrower traits or revise underwriting assumptions without qualified approval.

Proposed inputs

  1. Name

    Underwriting workbook copy

    Owner

    Authorized credit team

    Minimum scope

    Relevant sheets, formulas and cell references

    Quality check

    Preserve version and distinguish inputs from calculations

  2. Name

    Financial/policy evidence

    Owner

    Qualified reviewer

    Minimum scope

    Approved financial extracts, peer definitions and covenant rules

    Quality check

    Check periods, units and authority

Access and setup — proposed or unknown

Proposed minimum
  1. Use fictional financials and a workbook schema; no borrower account access, paid data retrieval or model edits.

Compatibility unknowns
  1. Current data-provider access, formula evaluation, workbook tooling and runtime compatibility are unknown.

External actions requiring approval
  1. Accessing or sharing financial information, changing models/memos and making credit commitments requires appropriate authorization.

Expected deliverable and review checks

Proposed artifact

Exception-brief candidate with cell pointers, source basis, missing assumptions and reviewer questions.

Critical checks
  1. Verify metric definitions, periods and formula lineage; separate policy exceptions from credit recommendations; retain uncertain evidence.

Missing output evidence

This overlay adds no new output inspection, execution or validation; consult the referenced canonical source outcome limits.

Failure signals and proposed responses

  1. Signal

    A ratio uses a different definition from the covenant

    Why it matters

    A comparison may flag the wrong exception

    Proposed response

    Align definitions or mark the test unresolved.

  2. Signal

    A pasted value replaces a formula

    Why it matters

    The model's lineage is obscured

    Proposed response

    Identify the cell and request the authoritative calculation.

Effort and cost — unknown

Setup work
  1. Define policy tests and workbook-reference conventions.

Ongoing review
  1. Qualified reviewers verify figures and own all decisions.

Unknowns
  1. Data rights, statement quality, calculation compatibility and professional-review effort are unknown.

First check — proposed, unexecuted

Status

proposed_unexecuted

Synthetic fixture

Invent a workbook excerpt with a period mismatch, unexplained override and covenant definition; prepare reference-linked questions only.

Acceptance checks
  1. Flag mismatches; preserve cell IDs; make no eligibility, credit or financial recommendation.

Stop conditions
  1. Stop if sensitive access is unapproved or a finding requires unsupported policy interpretation.

Limits

  1. A cell-linked brief cannot establish borrower creditworthiness, covenant compliance or the correctness of an underwriting decision.

Tasks: Money & accountingData & analysisSecurity & compliance

Useful for: Finance professional

Editorial relevance from the title and source-supported actions; not evidence of the operator’s occupation.
Claude CoworkIndependent testing: No

Editorial candidate. This is source-backed reporting or a documented workflow. BotCurrent has not independently tested this task. A proposed Canadian adaptation is not evidence of a Canadian deployment or compliance certification.

Original source

https://academy.claude.com/use-cases/draft-a-credit-memo-from-spreads-and-statements-with-claude-for-excel ↗

Attributed to: Anthropic

Official recipe
Original action type
tool_action
What the source supports
Retrieve financials and peer spreads, inspect the underwriting workbook for policy breaches and unsupported assumptions, and prepare a cell-referenced brief for reviewed model and memo updates.
Source outcome limits
This official recipe describes a procedure; it does not prove successful execution or independently reproduced results.
Evidence attribution caveat
This official recipe describes a procedure; it does not prove successful execution or independently reproduced results.
Source approval description
Human signs off on spread and owns credit decisions.
Named tools
Claude Cowork; S&P Global; Claude for Excel; Claude for Word
Model or version
Unknown
Publication date
Unknown
Source last checked
2026-10-14
Demonstrated in Canada
Unknown
Canadian availability
Unknown
Proposed Canadian adaptation
Apply the lender’s Canadian policies and CAD conventions; qualified staff alone make credit decisions.
Recommended approval boundaries
Review source references and artifacts; approve any posting, external send, ledger entry, contract acceptance or sensitive-data disclosure separately.