variance-commentary
Write flux commentary for every P&L and balance-sheet line over threshold — current vs prior period and vs budget, with the driver explained from underlying activity. Use for the month-end close package and management reporting.
Best for month-end close teamsWorks with GitHub
⌘source
- author
@anthropics- language
- Python
★11kstars
1.5kforks
May 6, 2026
✦overview.md
Key Features
- ·Threshold-based flagging with materiality and always-comment list
- ·Explanatory drivers from underlying activity, not number restatements
- ·Structured output: table plus narrative summary
- ·Integration with internal GL MCP for driver sourcing
- ·Fallback 'driver unclear' flag instead of invented explanations
Use Cases
- →Month-end close: automate commentary for P&L and balance sheet variance reports
- →Management reporting: produce consistent, data-backed commentary for board packs
- →Budget vs actuals analysis: explain movements beyond raw percentages
Best for
- ✓month-end close teams
- ✓FP&A analysts
- ✓management reporting
Not ideal for
- !ad-hoc one-off queries
- !non-financial data sets
FAQs
name
variance-commentary
description
Write flux commentary for every P&L and balance-sheet line over threshold — current vs prior period and vs budget, with the driver explained from underlying activity. Use for the month-end close package and management reporting.
Variance commentary
Given current-period actuals, prior-period actuals, and budget for the same scope, produce a commentary table.
Threshold
Flag a line for commentary if either is true:
- Absolute variance ≥ the firm's materiality threshold (use the provided value; default 5% of the line or a fixed floor, whichever is greater)
- The line is on the "always comment" list (revenue, headcount cost, cash)
For each flagged line
| Column | Content |
|---|---|
| Line | Account or caption |
| Current / Prior / Budget | The three values |
| Δ vs prior and Δ vs budget | Amount and % |
| Driver | One sentence explaining the movement from underlying activity — not a restatement of the number |
A driver explains why, not what: "Cloud spend up $1.2M on incremental GPU reservations for the May launch" — not "Cloud spend increased $1.2M (18%)."
Sourcing the driver
Look at the activity behind the line (journal-source breakdown, vendor mix, headcount delta, volume × rate) via the internal-gl MCP. If the driver isn't clear from the data, write "driver unclear — flag for controller" rather than inventing one.
Output
$install
1-click copynpx skills add anthropics/financial-services --skill variance-commentarySafety assessment
★
Clarity score
How clear and easy to understand the SKILL.md instructions are, rated from 1 to 5.
5/ 5
excellentVery clear and well structured, with almost no room for misunderstanding.
◎
Actionability score
How directly an agent can act on the SKILL.md instructions, rated from 1 to 5.
4/ 5
highMostly actionable with clear steps; only a few small gaps remain.