CFO
by
Weekly financial review for ecommerce teams: reconcile the numbers into one run rate, then name the risks and opportunities with what each is worth.
Most weekly financial reviews are a reconciliation exercise dressed as an analysis. Somebody pulls revenue from the storefront, spend from three ad platforms and cash from the bank, spends two hours making them agree, and presents a set of numbers with no decision attached. The interesting question, what is likely to go wrong in the next six weeks and what is it worth, never gets asked because the time went into the arithmetic. This skill reconciles the numbers into one run rate, then names the risks and opportunities with what each is worth.
What It Does
- Reconciles storefront, platform and bank figures into one set of numbers.
- Projects the run rate against the plan, with the drivers behind the gap.
- Tracks cash conversion and flags where it is stretching.
- Identifies the risk arriving in the next six weeks, before it is urgent.
- Sizes each opportunity in contribution rather than in revenue.
- Delivers the overview, the projection, and the risks and opportunities.
Use Cases
- Stop spending the weekly review reconciling numbers.
- See the cash squeeze six weeks out rather than the week it lands.
- Give the team a run rate everyone agrees on.
- Rank this week's opportunities by contribution.
Example Session
Give me the weekly financial picture and what I should be worried about.
Run rate is 4% under plan. The cash timing is the real issue.
- Item | This week | Run rate | vs Plan | Note
- Net revenue | $412,000 | $1.71M per month | -4% | Volume, not price
- Contribution margin | $164,000 | 39.8% | -1.2pts | Freight and returns
- Cash on hand | $1.24M | n/a | n/a | Q3 buy due in 5 weeks
- Days to cover | 47 | n/a | n/a | Tightest point week 6
The revenue gap is volume rather than discounting, which is the better of the two problems. Margin is down 1.2 points on freight and returns, worth about $20,000 a month. The thing to act on is week six: the Q3 buy lands while receivables are at their thinnest, and moving $300,000 of it out by two weeks removes the squeeze without touching anything else.
Why This Beats Prompting It Yourself
Numbers That Agree vs A Decision To Make: The Hidden Risk
A generic prompt produces a summary of the figures, which is the deliverable the request seems to ask for. Nothing in it says which week is going to be tight or what to do about it. The reconciliation is necessary work and it is not analysis, and a review that stops there is a status update with a spreadsheet attached.
When You Still Need The Week It Gets Tight
Cash problems are cheap six weeks ahead and expensive in the week they arrive. Naming week six, and the $300,000 of purchase order timing that resolves it, converts the entire review into one action. That number is invisible in any view that reports cash as a single current balance.
Known Limitations
Cash projection depends on payment terms and receivables timing being current, and stale terms make the picture look better than it is. Reconciliation resolves the differences it can explain and reports the residual rather than forcing agreement, so a persistent gap stays visible. It produces the review and does not move money or amend purchase orders.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/cfo/
SKILL.md
references/reconciliation.md
weekly/
# one file per week, with risks and opportunities
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from a finance dashboard?
A dashboard shows current state from whichever source each tile reads. This reconciles those sources into one run rate, reports what it could not reconcile rather than hiding it, and projects forward to name the week that needs action.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
What does it need connected to work?
Orders, ad spend by platform, product cost, bank balances, payment terms, and open purchase orders. Without purchase orders it still projects cash and says the inventory commitments are missing, which is the piece that usually causes the squeeze.
What does it do when the numbers do not reconcile?
It reports the residual with its best explanation rather than forcing agreement. A gap that persists week to week is itself a finding, usually a timing difference or a fee category nobody has mapped, and burying it in a rounding is how it survives for years.
Why project six weeks rather than to the quarter end?
Because six weeks is roughly the horizon where a purchase order can still be moved, a payment renegotiated or spend slowed. Further out is a forecast; six weeks is a decision window, and the skill also flags anything larger it can see beyond it.
What is included with the skill?
The skill itself, the reconciliation reference, the weekly review format, and the cash projection assumptions. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your projections start shifting.
A weekly financial read should not depend on which tab is open.
Polar reconciles revenue, margin and cash on one definition, so the run-rate and the risks you publish match the close.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from a finance dashboard?","acceptedAnswer":{"@type":"Answer","text":"A dashboard shows current state from whichever source each tile reads. This reconciles those sources into one run rate, reports what it could not reconcile rather than hiding it, and projects forward to name the week that needs action."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"Orders, ad spend by platform, product cost, bank balances, payment terms, and open purchase orders. Without purchase orders it still projects cash and says the inventory commitments are missing, which is the piece that usually causes the squeeze."}},{"@type":"Question","name":"What does it do when the numbers do not reconcile?","acceptedAnswer":{"@type":"Answer","text":"It reports the residual with its best explanation rather than forcing agreement. A gap that persists week to week is itself a finding, usually a timing difference or a fee category nobody has mapped, and burying it in a rounding is how it survives for years."}},{"@type":"Question","name":"Why project six weeks rather than to the quarter end?","acceptedAnswer":{"@type":"Answer","text":"Because six weeks is roughly the horizon where a purchase order can still be moved, a payment renegotiated or spend slowed. Further out is a forecast; six weeks is a decision window, and the skill also flags anything larger it can see beyond it."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the reconciliation reference, the weekly review format, and the cash projection assumptions. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your projections start shifting."}}]}</script>