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".

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

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.

A dense, well-structured audit playbook with concrete per-check tests, clear sequencing, and explicit priority gating. It is highly actionable and mostly concise; the only lever is optionally splitting the model-type bug catalogs into a reference file.

Suggestions

Move the model-type-specific bug lists (3g: DCF/LBO/Merger/3-statement) into a references/ file (e.g. MODEL_BUGS.md) and link to it from Step 3 to slim the main SKILL.md.

De-duplicate the BS-balance and hardcoded-override guidance between Step 3 and the Notes section to tighten token use.

Add a one-line note on how to access cell formulas (e.g. reading formula strings vs. values) so the formula-level checks are executable without assumed tooling.

Dimension	Reasoning	Score
Conciseness	Mostly lean: terse tables of checks/tests assume Claude's intelligence and avoid explaining what a balance sheet is. Minor repetition (BS balance and hardcoded-override warnings recur in Step 3 and Notes) keeps it just below fully tight.	4 / 5
Actionability	Concrete, executable tests throughout — e.g. 'Total Assets = Total Liabilities + Equity (every period)', 'CF Ending Cash = BS Cash', per-model bug lists (e.g. 'WACC uses book values instead of market values'). Guidance is specific enough to act on directly.	5 / 5
Workflow Clarity	Four clearly sequenced steps (scope → formula checks → model-integrity checks → report) with explicit gating checkpoints ('if BS doesn't balance ... nothing else matters until this is fixed') and checklists per model type; 'report first, fix on request' bounds the read-only workflow appropriately.	5 / 5
Progressive Disclosure	Well-organized into clearly headed sections with no nested references. As a single-file skill over ~150 lines it stays navigable, though the model-type-specific bug catalogs (3g) are candidates for an external reference that would tighten the main file.	4 / 5
	Total	18 / 20 Passed

Description

100%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, third-person description that concretely states capabilities, scopes, and integrity checks, and supplies a rich set of natural trigger phrases. Both the 'what' and 'when' are explicit and specific.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Audit a spreadsheet for formula accuracy, errors, and common mistakes', scope variants (range/sheet/model), and specific integrity checks ('BS balance, cash tie-out, and logic sanity') — giving comprehensive coverage.	5 / 5
Completeness	Explicitly answers both 'what' (audit formulas/errors/integrity checks across scopes) and 'when' (a dedicated 'Triggers on ...' clause with concrete phrases).	5 / 5
Trigger Term Quality	Ten natural trigger phrases a user would actually say ('audit this sheet', 'check my formulas', 'model won't balance', 'something's off in my model') cover synonyms and informal variants comprehensively.	5 / 5
Distinctiveness Conflict Risk	Targets a clear niche — spreadsheet/financial-model auditing — with distinct triggers unlikely to fire for unrelated skills.	5 / 5
	Total	20 / 20 Passed

Validation

100%

Warnings & errors only

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
Path: plugins/agent-plugins/earnings-reviewer/skills/audit-xls/SKILL.md
Commit: 69cbc81

Reviewed: 7 days ago

Table of Contents

Discovery Implementation Validation

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.