Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build the pipeline that turns team updates into the rollup report without the Friday scramble — the collection format that aggregates cleanly, the altitude translation (team detail → leadership signal), and the automation-lite assembly that takes minutes. Use when asked I compile status from five teams every week, streamline our reporting chain, my Friday is spent chasing updates, or make the rollup write itself. Produces the collection design, the translation rules, the assembly routine, and th
.claude/skills/mohitagw15856-status-report-pipeline/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 75% | 0% |
Somewhere in every org, a person spends Friday chasing five sub-team updates, translating them into one rollup, and reformatting everything because each team reports differently. The pipeline fixes the three leaks: collection (one shared format — async-update-format shape — due Thursday, in a place the assembler can harvest), translation (explicit altitude rules for what leadership needs: risks, asks, deltas — not activity), and assembly (a standing skeleton where harvested content drops into place in minutes). The chase disappears because the deadline-and-default replaces the reminder treadmill.
Ask for these if not provided:
The four-line input format · due: time] · harvest location · the no-update default line]
Ascends: 🔴🟡 + asks + notable deltas · compresses: the green line · never: activity · fidelity note on numbers]
The standing rollup structure · the harvest-and-place steps · the synthesis sentence last · target minutes]
The default mechanic · the two-cycle self-correction expectation · escalation belongs to (role), not the assembler]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 28,311 | 21,602 | -24% | 1 | 1 | 0% | 4,109 | 4,043 | -2% | 0 | 0 | — |
case-08 | fail→pass | 14,174 | 17,539 | +24% | 1 | 1 | 0% | 2,267 | 3,063 | +35% | 0 | 0 | — |
case-02 | fail→pass | 37,779 | 20,869 | -45% | 1 | 1 | 0% | 4,851 | 3,969 | -18% | 0 | 0 | — |
case-03 | fail→pass | 25,279 | 25,591 | +1% | 1 | 1 | 0% | 4,387 | 4,537 | +3% | 0 | 0 | — |
case-04 | fail→pass | 19,197 | 22,202 | +16% | 1 | 1 | 0% | 2,208 | 3,866 | +75% | 0 | 0 | — |
case-05 | pass→pass | 15,935 | 14,438 | -9% | 1 | 1 | 0% | 2,269 | 3,419 | +51% | 0 | 0 | — |
case-06 | fail→pass | 12,760 | 17,801 | +40% | 1 | 1 | 0% | 1,973 | 3,099 | +57% | 0 | 0 | — |
case-07 | fail→pass | 15,539 | 13,508 | -13% | 1 | 1 | 0% | 2,272 | 3,400 | +50% | 0 | 0 | — |
case-09 | pass→pass | 17,602 | 14,427 | -18% | 1 | 1 | 0% | 1,985 | 2,600 | +31% | 0 | 0 | — |
case-10 | pass→pass | 14,824 | 21,224 | +43% | 1 | 1 | 0% | 1,574 | 3,127 | +99% | 0 | 0 | — |
case-11 | pass→pass | 20,435 | 16,640 | -19% | 1 | 1 | 0% | 2,379 | 2,890 | +21% | 0 | 0 | — |
case-12 | pass→pass | 16,714 | 13,483 | -19% | 1 | 1 | 0% | 1,784 | 2,389 | +34% | 0 | 0 | — |
case-13 | fail→pass | 20,234 | 21,455 | +6% | 1 | 1 | 0% | 2,213 | 3,252 | +47% | 0 | 0 | — |
case-14 | pass→pass | 12,224 | 17,496 | +43% | 1 | 1 | 0% | 1,873 | 2,951 | +58% | 0 | 0 | — |
case-15 | pass→pass | 16,976 | 8,151 | -52% | 1 | 1 | 0% | 1,869 | 2,394 | +28% | 0 | 0 | — |
case-16 | fail→pass | 14,175 | 5,954 | -58% | 1 | 1 | 0% | 2,204 | 2,082 | -6% | 0 | 0 | — |
case-17 | fail→pass | 18,658 | 11,167 | -40% | 1 | 1 | 0% | 1,962 | 2,814 | +43% | 0 | 0 | — |
case-18 | pass→pass | 19,829 | 13,115 | -34% | 1 | 1 | 0% | 2,185 | 2,977 | +36% | 0 | 0 | — |
case-19 | pass→pass | 22,483 | 26,798 | +19% | 1 | 1 | 0% | 2,543 | 3,734 | +47% | 0 | 0 | — |
case-20 | pass→pass | 17,148 | 15,542 | -9% | 1 | 1 | 0% | 1,807 | 2,780 | +54% | 0 | 0 | — |
case-21 | fail→fail | 17,653 | 12,822 | -27% | 1 | 1 | 0% | 2,445 | 2,873 | +18% | 0 | 0 | — |
case-22 | pass→pass | 19,666 | 18,330 | -7% | 1 | 1 | 0% | 3,224 | 4,101 | +27% | 0 | 0 | — |
case-23 | pass→pass | 30,457 | 21,607 | -29% | 1 | 1 | 0% | 2,986 | 3,942 | +32% | 0 | 0 | — |
case-24 | fail→fail | 21,369 | 27,284 | +28% | 1 | 1 | 0% | 3,551 | 5,613 | +58% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +42 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.