Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a Performance Improvement Plan a manager can defend and an employee can actually act on — specific concerns, measurable goals, real support, and an honest timeline. Use when asked to write a PIP, put someone on a performance plan, document underperformance, or build a formal improvement plan. Produces the concern statement, the measurable goals with success criteria, the support plan, the check-in cadence, and the consequences — HR/legal-ready. Complements pip-responder (the employee's sid
.claude/skills/mohitagw15856-pip-writer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 104% | 0% |
A Performance Improvement Plan fails when it's vague ("improve communication"), unfair (goals no one could hit), or a paper trail dressed as help. This writes the version that is genuinely fair and defensible: concrete concerns tied to examples, goals with measurable success criteria, the support the company will provide, and honest stakes — so the outcome is either a real turnaround or a decision no one can call arbitrary.
> Not legal advice. Employment law varies by jurisdiction and contract — have HR/counsel review before delivery.
Ask for these if not provided:
Period: start]–end] · Manager: x] · HR partner: x]
| Concern | Specific example(s) | Expected standard | |---|---|---|
| Goal | Measurable success criterion | Evidence | By | |---|---|---|---|
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 21,363 | 18,827 | -12% | 1 | 1 | 0% | 2,853 | 3,243 | +14% | 0 | 0 | — |
case-02 | fail→pass | 23,100 | 11,969 | -48% | 1 | 1 | 0% | 2,748 | 3,029 | +10% | 0 | 0 | — |
case-03 | fail→pass | 20,108 | 20,224 | +1% | 1 | 1 | 0% | 2,758 | 3,680 | +33% | 0 | 0 | — |
case-04 | pass→pass | 21,968 | 18,822 | -14% | 1 | 1 | 0% | 2,437 | 3,089 | +27% | 0 | 0 | — |
case-05 | fail→pass | 23,108 | 19,755 | -15% | 1 | 1 | 0% | 2,332 | 3,369 | +44% | 0 | 0 | — |
case-06 | fail→pass | 8,218 | 18,636 | +127% | 1 | 1 | 0% | 1,538 | 3,131 | +104% | 0 | 0 | — |
case-07 | fail→pass | 21,451 | 17,222 | -20% | 1 | 1 | 0% | 2,900 | 2,913 | +0% | 0 | 0 | — |
case-08 | fail→pass | 15,535 | 18,438 | +19% | 1 | 1 | 0% | 1,819 | 3,188 | +75% | 0 | 0 | — |
case-09 | fail→fail | 22,520 | 20,231 | -10% | 1 | 1 | 0% | 2,683 | 3,232 | +20% | 0 | 0 | — |
case-10 | pass→pass | 18,854 | 17,763 | -6% | 1 | 1 | 0% | 2,374 | 3,032 | +28% | 0 | 0 | — |
case-11 | pass→pass | 22,825 | 17,394 | -24% | 1 | 1 | 0% | 2,865 | 3,811 | +33% | 0 | 0 | — |
case-12 | pass→pass | 21,439 | 23,457 | +9% | 1 | 1 | 0% | 2,696 | 3,399 | +26% | 0 | 0 | — |
case-13 | pass→pass | 17,094 | 20,657 | +21% | 1 | 1 | 0% | 1,850 | 3,102 | +68% | 0 | 0 | — |
case-14 | pass→pass | 21,536 | 20,068 | -7% | 1 | 1 | 0% | 2,388 | 3,299 | +38% | 0 | 0 | — |
case-15 | pass→pass | 24,525 | 22,422 | -9% | 1 | 1 | 0% | 2,912 | 3,321 | +14% | 0 | 0 | — |
case-16 | fail→pass | 23,547 | 19,487 | -17% | 1 | 1 | 0% | 2,756 | 3,083 | +12% | 0 | 0 | — |
case-17 | fail→pass | 20,447 | 23,344 | +14% | 1 | 1 | 0% | 2,385 | 3,698 | +55% | 0 | 0 | — |
case-18 | pass→pass | 16,013 | 15,490 | -3% | 1 | 1 | 0% | 1,673 | 2,901 | +73% | 0 | 0 | — |
case-19 | pass→pass | 20,599 | 18,012 | -13% | 1 | 1 | 0% | 2,612 | 2,986 | +14% | 0 | 0 | — |
case-20 | pass→fail | 21,529 | 18,582 | -14% | 1 | 1 | 0% | 2,384 | 2,958 | +24% | 0 | 0 | — |
case-21 | pass→pass | 12,988 | 20,108 | +55% | 1 | 1 | 0% | 1,929 | 2,791 | +45% | 0 | 0 | — |
case-22 | pass→pass | 21,331 | 14,110 | -34% | 1 | 1 | 0% | 2,554 | 2,963 | +16% | 0 | 0 | — |
case-23 | pass→pass | 20,504 | 24,087 | +17% | 1 | 1 | 0% | 2,200 | 3,402 | +55% | 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. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.