agentleFS
agiea-brainagiea-brain/agency-skills/skills/tapajyoti-das/cash-runway-forecast/SKILL.md

cash-runway-forecast

Skill1 starsChanged 17 days agoMITView on GitHub

Use this skill when the firm is profitable on paper and still worried about payroll, or before any commitment that spans a few months. Produces a thirteen-week view of money in and money out, the weeks that go tight, what causes each one, and the two or three moves that fix them while there is still time. Also use for "can we afford this hire", "will we make payroll", "cash flow forecast", "when do we get tight".

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 11 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start from the bank balance today, not from the profit and loss.

    line 19
  2. Distinctive:

    Run the balance week by week and circle every week it goes below the floor.

    line 27
  3. Distinctive:

    MUST start from the current bank balance, not from the profit and loss.

    line 45
  4. Distinctive:

    MUST date receivables by when each client actually pays, not by terms.

    line 46
  5. Distinctive:

    MUST include annual and quarterly outflows falling inside the window.

    line 47
  6. Distinctive:

    MUST define the cash floor before reading the output.

    line 48
  7. Distinctive:

    MUST name the cause of every week below the floor.

    line 49
  8. Distinctive:

    MUST order the fixes by reversibility.

    line 50
  9. Distinctive:

    MUST state the assumption the forecast most depends on.

    line 51
  10. Distinctive:

    NEVER project revenue forward in a straight line.

    line 52
  11. Distinctive:

    NEVER present a forecast with no tight weeks without saying why there are none.

    line 53

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Applies to a firm whose profit and loss looks fine and whose bank balance moves unpredictably. Produces the thirteen-week forecast and the decisions it surfaces.

The play

  1. Start from the bank balance today, not from the profit and loss. Profit is an opinion about timing and cash is a fact, and a firm can be profitable every month of a year it runs out of money in.
  2. Lay out thirteen weeks because that is far enough to act and near enough to be roughly right. Beyond a quarter the numbers are a plan, not a forecast.
  3. Put money in by the week it will actually arrive, which means the invoice date plus terms plus how late that client is in practice. A client who has paid at day fifty every time for a year does not pay at day thirty in this forecast.
  4. Put money out by the week it leaves: payroll and its dates, contractors, tax, software, rent, anything annual that lands in one of these thirteen weeks. The annual items are what catch firms out, because they are not in the monthly rhythm anyone carries in their head.
  5. Run the balance week by week and circle every week it goes below the floor. The floor is not zero. It is the amount the firm needs to not be making decisions under pressure, and it should be written down before the forecast is read.
  6. For each tight week, name the cause. It is almost always one of three: a large invoice landing late, two payroll runs in one month, or an annual bill nobody diaried. The cause determines the fix, and the fixes are different.
  7. Give the moves in order of reversibility: accelerate a receivable, move a discretionary payment, delay a commitment, draw on credit, and only then anything structural. A tight week identified eight weeks out is an email. The same week found three days out is a loan.
  8. Say what the forecast assumes and which assumption breaks it. Usually one client's payment timing carries the whole thing, and knowing which one tells the firm what to watch.

What good looks like

  • First thing I check that most people skip: the annual and quarterly bills landing inside the window. Insurance, tax, the software renewal that bills once a year. They are invisible in a monthly mental model and they are what turns an ordinary week into a tight one.
  • The mediocre version: last month's revenue projected forward in a straight line. It produces a smooth chart with no tight weeks in it, which is the one result a cash forecast should never produce.
  • I know it is right when the receivables are dated by when each client actually pays rather than by the terms on the invoice.
  • Second test: the floor was written down before anyone looked at the output, so the tight weeks were not defined after the fact to make the picture acceptable.
  • Red flag: no week goes below the floor. Either the floor is too low or money in has been entered optimistically, and both mean the forecast will not warn anybody about anything.

Rules

  • MUST start from the current bank balance, not from the profit and loss.
  • MUST date receivables by when each client actually pays, not by terms.
  • MUST include annual and quarterly outflows falling inside the window.
  • MUST define the cash floor before reading the output.
  • MUST name the cause of every week below the floor.
  • MUST order the fixes by reversibility.
  • MUST state the assumption the forecast most depends on.
  • NEVER project revenue forward in a straight line.
  • NEVER present a forecast with no tight weeks without saying why there are none.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.