Use when a support account wobbles or a quarterly business review looms: read the health signals, run the success plan, and prep the review from evidence.
The English summary is being prepared. Showing the original description.
Use when the user asks about budgets, pacing against plan, forecasts, re-forecasts, annual plans, or where a period lands at the current run rate.
The English summary is being prepared. Showing the original description.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
Run this when the user wants to know where a period lands: pacing against plan, the forecast behind it, or a re-forecast after something moved. Read from the ledger only — anything untraceable is UNKNOWN, never estimated.
The finance ledger, the budget set with one row per line and owner, and the period calendar.
Forecast driver-first: name the 5-15 drivers that actually move the plan — units times price, headcount times rate, pipeline times win rate — and give each one an owner in the assumption log. Model the drivers, then let the dollars follow. A plan built straight from last year plus a percentage is not a forecast.
Per line: actuals to date, the share of the plan already used, days elapsed against days in the period, and where the period lands if the current run rate holds. Show the math, not just the verdict. Keep a rolling 12-18 month horizon — drop the expired month, add a new one at the end.
This is also the pace check the user can ask for at any point in the month: every budget line read at once, with month-to-date actuals and the landing point per line.
Flag any line projected to pass its plan, and give the run rate it would take to land on plan instead. Flag any line pacing more than 15% under plan too — budget left on the table is a miss, not a saving. Every flag carries the gap in currency, the driver behind it, and one play with a named owner.
Base (commit), Adverse (downside), or Opportunity (upside), with one evidence line behind each grade. Base needs a named driver, not hope.
Re-forecast monthly with a deeper refresh each quarter. On every re-forecast: restate the prior forecast, name what moved and by how much, and carry forward only the drivers still true. Never silently re-baseline. Track accuracy against the prior forecast — bias direction and miss size — and feed repeat misses back into the driver assumptions.
Move named money between lines with the total flat, cap any single move at 30% of the source line, and say which owner approved each move. Draft only — you never move a budget yourself.
If a month or quarter is missing from the ledger, say which one and ask for that export. Do not smooth over the gap with a projection, and do not put a placeholder number where an actual belongs.
One line per budget line, then the total, then the forecast grade and any reallocation draft. Under 200 words unless they asked for a table.
No budget set: build the read against last period's actuals and label every plan number INFERENCE. No ledger at all: ask for one export and say exactly which columns you need.
Reads and drafts only. Budget moves, plan changes, and anything leaving the team wait for the owner's yes.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Use when a support account wobbles or a quarterly business review looms: read the health signals, run the success plan, and prep the review from evidence.
The English summary is being prepared. Showing the original description.
Reconcile accounts by comparing GL balances to subledgers, bank statements, or third-party data. Use when performing bank reconciliations, GL-to-subledger recs, intercompany reconciliations, or identifying and categorizing reconciling items.
The English summary is being prepared. Showing the original description.
Review open receivables and draft factual, staged payment follow-ups without inventing status or contacting a customer.
The English summary is being prepared. Showing the original description.
Write ad variants that each test one idea, within platform limits and supported claims.
The English summary is being prepared. Showing the original description.
Use on the first conversation with Alex, or whenever their memory has no product preferences yet: learn what the user is building and who for, where specs, roadmap and numbers live, who decides dates and scope, and get them to a first real product deliverable.
The English summary is being prepared. Showing the original description.
Use when a strategic alliance needs joint selling governance: operating model, joint targeting, steering prep, and milestone accountability.
The English summary is being prepared. Showing the original description.