sox-testing

tarafından anthropic

SOX örnek seçimleri, test çalışma kağıtları ve kontrol değerlendirmeleri oluşturun. Üç aylık veya yıllık SOX 404 testi planlaması yaparken, bir örneklem çekerken kullanın…

npx skills add https://github.com/anthropics/knowledge-work-plugins --skill sox-testing

SOX Compliance Testing

If you see unfamiliar placeholders or need to check which tools are connected, see CONNECTORS.md.

Important: This command assists with SOX compliance workflows but does not provide audit or legal advice. All testing workpapers and assessments should be reviewed by qualified financial professionals before use in audit documentation.

Generate sample selections, create testing workpapers, document control assessments, and provide testing templates for SOX 404 internal controls over financial reporting.

Usage

/sox <control-area> <period>

Arguments

  • control-area — The control area to test:
    • revenue-recognition — Revenue cycle controls (order-to-cash)
    • procure-to-pay or p2p — Procurement and AP controls (purchase-to-pay)
    • payroll — Payroll processing and compensation controls
    • financial-close — Period-end close and reporting controls
    • treasury — Cash management and treasury controls
    • fixed-assets — Capital asset lifecycle controls
    • inventory — Inventory valuation and management controls
    • itgc — IT general controls (access, change management, operations)
    • entity-level — Entity-level and monitoring controls
    • journal-entries — Journal entry processing controls
    • Any specific control ID or name
  • period — The testing period (e.g., 2024-Q4, 2024, 2024-H2)

Workflow

1. Identify Controls to Test

Based on the control area, identify the key controls. Present the control matrix:

Control #Control DescriptionTypeFrequencyKey/Non-KeyRiskAssertion
[ID][Description]Manual/Automated/IT-DependentDaily/Weekly/Monthly/Quarterly/AnnualKeyHigh/Medium/Low[CEAVOP]

Control types:

  • Automated: System-enforced controls with no manual intervention
  • Manual: Controls performed by personnel with judgment
  • IT-dependent manual: Manual controls that rely on system-generated data

Assertions (CEAVOP):

  • Completeness — All transactions are recorded
  • Existence/Occurrence — Transactions actually occurred
  • Accuracy — Amounts are correctly recorded
  • Valuation — Assets/liabilities are properly valued
  • Obligations/Rights — Entity has rights to assets, obligations for liabilities
  • Presentation/Disclosure — Properly classified and disclosed

2. Determine Sample Size

Calculate sample sizes based on control frequency and risk:

Control FrequencyPopulation Size (approx.)Recommended Sample
Annual11 (test the instance)
Quarterly42
Monthly122-4 (based on risk)
Weekly525-15 (based on risk)
Daily~25020-40 (based on risk)
Per-transactionVaries25-60 (based on risk and volume)

Adjust for:

  • Risk level: Higher risk controls require larger samples
  • Prior year results: Controls with prior deficiencies need larger samples
  • Reliance: Controls relied upon by external auditors may need larger samples

3. Generate Sample Selection

Select samples from the population using the appropriate method:

Random selection (default for transaction-level controls):

  • Generate random numbers to select specific items from the population
  • Ensure coverage across the full period

Systematic selection (for periodic controls):

  • Select items at fixed intervals with a random start point
  • Ensure representation across all sub-periods

Targeted selection (supplement to random, for risk-based testing):

  • Select items with specific risk characteristics (high dollar, unusual, period-end)
  • Document rationale for targeted selections

Present the sample:

SAMPLE SELECTION
Control: [Control ID] — [Description]
Period: [Testing period]
Population: [Count] items, $[Total value]
Sample size: [N] items
Selection method: [Random/Systematic/Targeted]

| Sample # | Transaction Date | Reference/ID | Amount | Selection Basis |
|----------|-----------------|--------------|--------|-----------------|
| 1        | [Date]          | [Ref]        | $X,XXX | Random          |
| 2        | [Date]          | [Ref]        | $X,XXX | Random          |
| ...      | ...             | ...          | ...    | ...             |

4. Create Testing Workpaper

Generate a testing template for each control:

SOX CONTROL TESTING WORKPAPER
==============================
Control #: [ID]
Control Description: [Full description of the control activity]
Control Owner: [Role/title — to be filled by tester]
Control Type: [Manual/Automated/IT-Dependent Manual]
Frequency: [How often the control operates]
Key Control: [Yes/No]
Relevant Assertion(s): [CEAVOP]
Testing Period: [Period]

TEST OBJECTIVE:
To determine whether [control description] operated effectively throughout the testing period.

TEST PROCEDURES:
1. [Step 1 — What to inspect, examine, or re-perform]
2. [Step 2 — What evidence to obtain]
3. [Step 3 — What to compare or verify]
4. [Step 4 — How to evaluate completeness of performance]
5. [Step 5 — How to assess timeliness of performance]

EXPECTED EVIDENCE:
- [Document type 1 — e.g., signed approval form]
- [Document type 2 — e.g., system screenshot showing review]
- [Document type 3 — e.g., reconciliation with preparer sign-off]

TEST RESULTS:

| Sample # | Ref | Procedure 1 | Procedure 2 | Procedure 3 | Result | Exception? | Notes |
|----------|-----|-------------|-------------|-------------|--------|------------|-------|
| 1        |     | Pass/Fail   | Pass/Fail   | Pass/Fail   | Pass/Fail | Y/N    |       |
| 2        |     | Pass/Fail   | Pass/Fail   | Pass/Fail   | Pass/Fail | Y/N    |       |

EXCEPTIONS NOTED:
| Sample # | Exception Description | Root Cause | Compensating Control | Impact |
|----------|----------------------|------------|---------------------|--------|
|          |                      |            |                     |        |

CONCLUSION:
[ ] Effective — Control operated effectively with no exceptions
[ ] Effective with exceptions — Control operated effectively; exceptions are isolated
[ ] Deficiency — Control did not operate effectively
[ ] Significant Deficiency — Deficiency is more than inconsequential
[ ] Material Weakness — Reasonable possibility of material misstatement not prevented/detected

Tested by: ________________  Date: ________
Reviewed by: _______________  Date: ________

5. Provide Common Control Templates

Based on the control area, provide pre-built test step templates:

Revenue Recognition:

  • Verify sales order approval and authorization
  • Confirm delivery/performance evidence
  • Test revenue recognition timing against contract terms
  • Verify pricing accuracy to contract/price list
  • Test credit memo approval and validity

Procure to Pay:

  • Verify purchase order approval and authorization limits
  • Confirm three-way match (PO, receipt, invoice)
  • Test vendor master data change controls
  • Verify payment approval and segregation of duties
  • Test duplicate payment prevention controls

Financial Close:

  • Verify account reconciliation completeness and timeliness
  • Test journal entry approval and segregation of duties
  • Verify management review of financial statements
  • Test consolidation and elimination entries
  • Verify disclosure checklist completion

ITGC:

  • Test user access provisioning and de-provisioning
  • Verify privileged access reviews
  • Test change management approval and testing
  • Verify batch job monitoring and exception handling
  • Test backup and recovery procedures

6. Document Control Assessment

Classify any identified deficiencies:

Deficiency: A control does not allow management or employees to prevent or detect misstatements on a timely basis. Consider:

  • Likelihood of misstatement
  • Magnitude of potential misstatement
  • Whether compensating controls exist

Significant Deficiency: A deficiency (or combination) that is less severe than a material weakness but important enough to merit attention by those responsible for oversight.

Material Weakness: A deficiency (or combination) such that there is a reasonable possibility that a material misstatement will not be prevented or detected on a timely basis.

7. Output

Provide:

  1. Control matrix for the selected area
  2. Sample selections with methodology documentation
  3. Testing workpaper templates with pre-populated test steps
  4. Results documentation template
  5. Deficiency evaluation framework (if exceptions are identified)
  6. Suggested remediation actions for any noted deficiencies

anthropic tarafından daha fazla skill

analyzing-financial-statements
anthropic
Bu beceri, yatırım analizi için finansal tablo verilerinden temel finansal oranları ve metrikleri hesaplar.
applying-brand-guidelines
anthropic
Bu beceri, oluşturulan tüm belgelere renkler, yazı tipleri, düzenler ve mesajlaşma dahil olmak üzere tutarlı kurumsal markalama ve stil uygular.
creating-financial-models
anthropic
Bu beceri, DCF analizi, duyarlılık testi, Monte Carlo simülasyonları ve yatırım için senaryo planlaması ile gelişmiş bir finansal modelleme paketi sunar…
board-minutes
anthropic
Kurul veya komite toplantı tutanaklarını ev formatınızda taslak haline getirir. Takviminizden yaklaşan kurul ve komite toplantılarını otomatik olarak algılar, gündemi sorar ve…
crm-cleanup
anthropic
HubSpot'u bayat fırsatlar, yinelenen kişiler ve eksik alanlar için tarar, ardından sahibinin onayladıklarını düzeltir. Fırsatlar, kişiler için isteğe bağlı kapsam argümanını kabul eder,…
redshift-api
anthropic
Amazon Redshift üzerinde SQL çalıştırın — ifadeler gönderin, durumu sorgulayın, sonuçlar arasında gezinin ve veritabanları/şemalar/tablolar arasında göz atın. Kullanıcı istediğinde bunu kullanın…
ticket-deflector
anthropic
Yönlendirilen bir müşteri e-postasını veya talebini okur, PayPal'dan sipariş/iade durumunu ve HubSpot'tan hesap geçmişini çeker, sahibinin üslubuna uygun bir yanıt taslağı oluşturur…
reg-feed-watcher
anthropic
Düzenleyici beslemeleri şimdi kontrol et ve son kontrolden bu yana yeni olanları, maddilik eşiğine göre filtreleyerek raporla. Kullanıcı "beslemeleri kontrol et" dediğinde kullanılır,…