Weekly Client Report

Before you start, ask me about each [input] below — one question at a time, with ready-made answers to pick from; skip whatever you can already tell.

Write the weekly performance report for [client: if not listed, call roasr_list_clients and let me pick], to be read by [reader: the business owner, non-technical (default) | the marketing manager | my internal team]. Run it [schedule: every Monday 9:00 (default) | every Friday 9:00 | first day of the month | custom].

1. **Load the RoASr method.** Call roasr_load_skill('roasr_frame'), roasr_load_skill('report_format') and roasr_load_skill('profit_math') and follow them exactly.
2. **Frame the data.** Call roasr_limits, roasr_list_clients and roasr_list_accounts: plan, niche, margin, KPI, account currencies, sync status and dataSince. The period is [period: last full week, Monday to Sunday (default) | last 7 days | custom], compared with the period of equal length right before it. If a window starts before dataSince, say so and narrow it. If currency.mixed is true, never sum money across accounts: report money per account. Name any account with a failed or never sync and draw no conclusions from it.
3. **Check for a previous report.** If I pasted last week's report or you have it in this conversation, use it for deltas. Otherwise this is the first run: write a short baseline and mark the comparison as "vs the prior period only".
4. **Headline KPIs.** roasr_metrics_summary for both periods: spend, conversions, revenue, ROAS, CPA, CTR, CPC. Show change in absolute and percent. Treat a move as real only above the noise floor: under about 30 conversions or 1,000 clicks in a period, call differences "within noise".
5. **Shape of the week.** roasr_metrics_daily for the current period: best and worst day, any day that looks like a tracking gap (spend with zero conversions). Do not add up reach or frequency across days.
6. **Campaigns.** roasr_metrics_campaigns for both periods, campaign level, limit 30. Pick the top 3 and bottom 3 by ROAS (or CPA for lead gen), only among campaigns with at least 5% of spend. Classify by configuration and data, not by name. Under a campaign budget, ad set numbers show allocation, not a ranking.
7. **Findings.** roasr_list_findings for this client: high-severity findings that are new this period, and ones that were closed.
8. **Profit.** Only if the client has a margin: profit = revenue × margin − spend, and compare ROAS with break-even ROAS = 1 / margin. No margin: ROAS and CPA only, and say once that profit is not shown.
9. **What was done.** Take it only from what I tell you and from findings that changed status. Never invent work that was not done.

**Delta discipline:** the first run records the baseline. Later runs report only what changed since the previous report: KPIs that crossed the noise floor, campaigns that entered or left the top/bottom, new or closed high findings. If nothing material changed, reply exactly: Nothing important has changed since the previous run. — and stop.

**Output:** a report that can go to the reader as is. One headline sentence with the week's main result in numbers. Then a KPI table (metric · this period · previous · change), a table of the top 3 and bottom 3 campaigns, then "What was done" and "What we recommend next" with 3–6 actions grouped High / Medium / Nice-to-have, each with the expected effect. End with caveats from the frame.

**Output rules:**
- Recommendations only — never apply changes; RoASr tools are read-only.
- Numbers over adjectives; every claim carries its figure and period.
- If the client has no margin set, talk ROAS and CPA, not profit.
- Keep money per account when currencies are mixed.
- Answer in the language of this prompt.