CtrlK
BlogDocsLog inGet started
Tessl Logo

audit-xls

Audit a spreadsheet for formula accuracy, errors, and common mistakes. Scopes to a selected range, a single sheet, or the entire model (including financial-model integrity checks like BS balance, cash tie-out, and logic sanity). Triggers on "audit this sheet", "check my formulas", "find formula errors", "QA this spreadsheet", "sanity check this", "debug model", "model check", "model won't balance", "something's off in my model", "model review".

72

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

The canonical home for this skill is audit-xls in anthropics/financial-services

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An expert-level audit checklist that treats the context window respectfully and encodes genuine financial-modeling QA knowledge as concrete, testable invariants with a clear reporting format. The main improvement levers are specifying how to mechanically open and scan the spreadsheet, and moving per-model bug catalogs into reference files as the skill grows.

Suggestions

Add a concrete mechanism for reading the spreadsheet (e.g., an openpyxl/xlsx script or a named tool) so the inspection step is as executable as the checks themselves.

Move the model-type-specific bug lists (DCF/LBO/merger/3-statement) into per-model reference files under references/ and link them from Step 3g, keeping the main body focused on the universal workflow.

Include a short worked example of the findings table with one filled-in row so the report format is unambiguous on first use.

DimensionReasoningScore

Conciseness

The body is dense, well-assumed-intelligent checklist content — it never explains what a balance sheet or DCF is, and every line ('PIK interest not accruing to principal', 'Terminal value not discounted back') is domain knowledge Claude would otherwise have to rediscover. Matches the 'lean and efficient; every token earns its place' anchor.

5 / 5

Actionability

Concrete, executable checks with explicit equality tests ('Total Assets = Total Liabilities + Equity (every period)', 'CFO + CFI + CFF = Δ Cash') plus a fully specified output table with columns and severity definitions. Falls short of anchor 5 because there is no mechanism for actually reading or parsing the spreadsheet file — no script, tool, or library pointer — leaving the inspection step to be improvised.

4 / 5

Workflow Clarity

A clear four-step sequence (determine scope → formula-level checks → model-integrity checks → report) where the steps themselves are validation checkpoints, augmented with triage guidance ('BS balance first', 'quantify the gap per period and trace where it breaks') and an explicit report-before-fix control ('Don't change anything without asking'). Matches the checklist-and-checkpoint anchor.

5 / 5

Progressive Disclosure

Well-organized sectioned structure with no bundle files, so all content lives appropriately in one SKILL.md. The model-type-specific bug lists (DCF / LBO / merger / 3-statement, ~30 lines) are the one block that could be split into per-model reference files to keep the main file tighter, keeping this below anchor 5.

4 / 5

Total

18

/

20

Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with explicit what-and-when structure, concrete capability examples, and an unusually rich set of natural trigger phrases. The only gaps are missing common synonyms and file extensions (Excel, workbook, .xlsx) that would push trigger coverage to fully comprehensive.

DimensionReasoningScore

Specificity

The description names the domain and several concrete actions — 'Audit a spreadsheet for formula accuracy, errors, and common mistakes', scoped to 'a selected range, a single sheet, or the entire model', with concrete integrity checks like 'BS balance, cash tie-out, and logic sanity'. It stops short of anchor 5 because it doesn't enumerate the full check coverage (formula errors, hardcodes, cross-sheet links) that the body delivers.

4 / 5

Completeness

Clearly and explicitly answers both: 'what' (audit a spreadsheet for formula accuracy, errors, and common mistakes, with three defined scopes) and 'when' via an explicit 'Triggers on' clause listing eight concrete trigger phrases. Matches the anchor-5 good example structure.

5 / 5

Trigger Term Quality

Strong natural phrases users would actually say: 'audit this sheet', 'check my formulas', 'QA this spreadsheet', 'model won't balance', "something's off in my model". Falls short of anchor 5 because common synonyms and file extensions like 'Excel', 'workbook', or '.xlsx' are absent.

4 / 5

Distinctiveness Conflict Risk

Clear niche — spreadsheet/financial-model auditing — with distinct, domain-specific triggers ('debug model', 'model won't balance') that would not fire for generic spreadsheet-creation or document skills. Minimal conflict risk.

5 / 5

Total

18

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
anthropics/financial-services
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.