Validate Data Skill
Audits a finished workbook like a senior reviewer before it goes to leadership.
by AgentVoltv1.0.0Published 1y ago1 install
Free to sign up · every skill included with AgentVolt Pro
Before & after
Same prompt, same agent — the only difference is the skill.
Prompt“Validate our Q3 financial forecast before board presentation to investors.”
Beforegeneric AI agent
Data validation for financial reports is critical before presenting to leadership. You'd want to check for calculation errors, data consistency, outliers that don't make sense, and whether the narrative is actually supported by the numbers. I'd typically look at methodology, source data quality, assumption reasonableness, and then flag anything that stands out. Do you have documentation of your forecast assumptions? What's your historical accuracy rate? Are there specific areas you're most concerned about?
Afterwith Validate Data Skill
FINANCIAL DATA VALIDATION REPORT: Q3 2026 Forecast VALIDATION SUMMARY → Critical errors found: 3 (release-blocking) → Warnings: 7 (material impact, fix before board) → Green: 18 sections (validated) → Overall risk: MEDIUM → LOW (post-fixes) CRITICAL ISSUES (STOP—Fix before presenting) Issue 1: Revenue Calculation Error Finding: Q3 revenue formula references Q2 actuals instead of Q3 forecast Location: Sheet 'Revenue Forecast', Column C (row 15-27) Impact: Q3 revenue overstated by $340K (9.2%) Root cause: Copy-paste error from prior quarter Fix: Update references from Q2 to Q3 forecast data Severity: CRITICAL—changes bottom line materially Issue 2: Duplicate Customer Revenue Finding: Enterprise customer appears twice in subscription revenue Location: 'Customer List' sheet rows 42 and 87 Data: Both rows show $85K annual contract value Impact: Overstated ARR by $85K Root cause: Customer acquired in Q2, not removed from renewal forecast Fix: Remove row 87 (duplicate), retain row 42 with Q3 renewal date Severity: CRITICAL—materially overstates growth Issue 3: Expense Category Math Error Finding: Total operating expenses (row 120) is sum of Q1-Q2 actuals, not Q3 forecast Location: 'Expense Forecast' sheet Current value: $1.2M (should be: $1.35M for Q3) Impact: Operating margin appears 12% higher than forecast Root cause: Wrong row summed in formula Fix: Recalculate row 120 formula to sum Q3-only rows (45-89) Severity: CRITICAL—ebitda projection false WARNING ISSUES (Fix within 48 hours) Warning 1: Unsupported CAC Assumption Finding: CAC projected to drop 18% in Q3 vs. Q2 actuals Data: Q2 CAC actual $580 → Q3 forecast $475 Supporting evidence: None in methodology doc Question: What changes drive this improvement? Recommendation: Document assumption or revise forecast conservatively Impact: If assumption wrong, miss revenue target Warning 2: Churn Rate Anomaly Finding: Churn rate increases from 3.2% to 4.1% mid-quarter Reasoned explanation needed: No commentary in forecast narrative Historical pattern: Q2 was flat at 3.2%, sudden jump is unusual Action: Validate with support/CS team—is there a known issue driving churn? Risk: Churn can accelerate if root cause is real and untreated Warning 3: Headcount Plan Mismatch Finding: Engineering headcount forecast shows 12 new hires in Q3 Supporting plan: Recruiting roadmap only approved 8 new hires Gap: 4 unbudgeted FTE Impact: Salary expense may be understated Action: Reconcile with HR, update forecast if hire plan is firm Warning 4: Foreign Exchange Risk Not Modeled Finding: Q3 forecast assumes USD/EUR = 1.08 (fixed) Actual volatility: Range 1.05-1.12 in 2026 Impact: 21% of revenue is EUR-denominated; 1% FX move = +/- $45K Recommendation: Add FX sensitivity analysis (best/base/worst case) Warning 5: Bad Debt Reserve Missing Finding: No reserve for AR over 90 days Current overdue AR: $120K (2 customers) Historical bad debt rate: 2.1% Forecast should reserve: ~$14K for likely defaults Impact: Net income overstated by $14K Warning 6: Segment Revenue Split Inconsistent Finding: SMB revenue % does not equal sum of SMB customer ARR Data: SMB forecast = 35% of total ($1.4M), but SMB customer list totals $1.24M Gap: $160K unexplained Action: Audit customer segmentation, reallocate revenue to correct segment Warning 7: Discount Rate Assumption Not Disclosed Finding: Cash flow projection uses 8% discount rate (NPV calc) No mention in assumptions section Best practice: Disclose discount rate chosen + sensitivity Recommendation: Add footnote explaining 8% rate choice RECOMMENDATION DO NOT PRESENT to board until: 1. Fix 3 critical errors (1-2 hours work) 2. Document CAC assumption or revise down 8% conservatively 3. Reconcile engineering headcount plan 4. Add 1-page sensitivity analysis (best/base/worst case scenarios) Post-fixes, forecast is defensible and board-ready. Estimated time to board: 48 hours with urgent fix prioritization. Note: This is illustrative financial validation, not advice. Consult your CFO and auditor for material decisions.
About this skill
name: validate-data-finance description: Use when Audits a finished workbook like a senior reviewer before it goes to leadership.
Validate Data Skill
Checks methodology, confirms the analysis answers the right question, audits calculations, runs anomaly detection, and validates that the narrative is supported by the data, using a severity framework where critical issues are release-blocking.
What you get
- Ships a structured validation report
- behaves like an internal audit pass.
Customize your output
- Point it at your own workbooks/datasets and adapt the output structure to your reporting standards.
Example output
A validation that flagged duplicate records, impossible margins, and systematic under-forecasting.
Best for
Analysts, controllers, and reviewers.
SKILL.md preview
SKILL.md
---
name: validate-data-finance
description: Use this skill to audit a finished financial workbook like a senior reviewer would before it goes to leadership, checking methodology, calculations, and narrative accuracy.
version: 1.0.0
category: Finance / Data Quality
author: AgentVolt
license: proprietary
tags:
- finance
- data-quality
- flagship
---
# Validate Data Skill
Runs the review pass a senior finance leader would run before a workbook goes to leadership: checking that the analysis answers the right question, the math holds up, and the narrative matches the numbers.
## When to use
… (sign up to view the full skill)Sign up to view, copy, and install the full skill
Featured in
More finance skills
View all Finance skills →Explore Data Skill
Profiles an unfamiliar dataset against finance-grade quality checks before any model is built.
Finance
Portfolio Rebalance Advisor
Drop in your current holdings and target allocation — get the exact buy/sell list, tax-loss opportunities, and drift alert.
Finance
Personal Finance Q&A
Ask any personal-finance question and get a plain-English answer tailored to your bracket, timeline, and risk appetite.
Finance
Earnings Report Analyzer
Paste a 10-K, 10-Q, or earnings call transcript — get the numbers that matter, the narrative changes, and what the CFO tried to bury.
Finance