Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a clear, inclusive, and structured job description for any role. Use when asked to write a job description, job posting, JD, or job advert. Produces a complete JD with role summary, responsibilities, requirements, and inclusive language review.
.claude/skills/mohitagw15856-job-description-writer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 13% | 0% |
Writes complete, inclusive job descriptions that attract the right candidates and reduce bias in the hiring process.
Company] | Location] | Remote policy] | Salary range]
About Company] 2-3 sentences. Specific and honest — not marketing copy.]
The Role 3-4 sentences. What this person will own, why the role exists now, what success looks like in year one.]
What You Will Do 6-8 bullet points. Outcomes and responsibilities, not activities. Start each with an action verb. Most important first.]
What We Are Looking For
Must have (4-6 items only):
Nice to have (3-4 items):
What We Offer Compensation, benefits, development. Be specific.]
How to Apply Clear instructions. What to send, where, timeline.]
Words to remove or replace:
| Original | Replace with | Why | |---|---|---| | "rockstar" | "experienced" | Gendered connotation | | "ninja" | "skilled" | Same issue | | "must have degree" | "relevant experience or qualification" | Excludes qualified non-graduates |
Requirement audit:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 14,379 | 14,577 | +1% | 1 | 1 | 0% | 2,745 | 3,523 | +28% | 0 | 0 | — |
case-02 | fail→pass | 15,091 | 14,056 | -7% | 1 | 1 | 0% | 2,893 | 3,235 | +12% | 0 | 0 | — |
case-03 | pass→pass | 15,603 | 12,465 | -20% | 1 | 1 | 0% | 2,824 | 3,018 | +7% | 0 | 0 | — |
case-04 | pass→pass | 15,792 | 14,965 | -5% | 1 | 1 | 0% | 2,737 | 3,111 | +14% | 0 | 0 | — |
case-05 | pass→pass | 13,443 | 11,336 | -16% | 1 | 1 | 0% | 2,486 | 2,603 | +5% | 0 | 0 | — |
case-06 | pass→pass | 14,251 | 10,978 | -23% | 1 | 1 | 0% | 2,310 | 2,588 | +12% | 0 | 0 | — |
case-07 | fail→pass | 12,807 | 10,349 | -19% | 1 | 1 | 0% | 2,202 | 2,352 | +7% | 0 | 0 | — |
case-08 | fail→pass | 13,581 | 13,068 | -4% | 1 | 1 | 0% | 2,412 | 3,016 | +25% | 0 | 0 | — |
case-09 | fail→pass | 12,989 | 10,709 | -18% | 1 | 1 | 0% | 2,332 | 2,630 | +13% | 0 | 0 | — |
case-10 | fail→fail | 11,455 | 16,273 | +42% | 1 | 1 | 0% | 2,079 | 3,493 | +68% | 0 | 0 | — |
case-11 | fail→pass | 10,870 | 11,906 | +10% | 1 | 1 | 0% | 2,054 | 2,958 | +44% | 0 | 0 | — |
case-12 | fail→pass | 10,949 | 13,260 | +21% | 1 | 1 | 0% | 1,999 | 2,955 | +48% | 0 | 0 | — |
case-13 | fail→pass | 13,910 | 13,483 | -3% | 1 | 1 | 0% | 2,611 | 3,169 | +21% | 0 | 0 | — |
case-14 | fail→pass | 13,518 | 15,245 | +13% | 1 | 1 | 0% | 2,483 | 3,453 | +39% | 0 | 0 | — |
case-19 | fail→pass | 9,309 | 12,080 | +30% | 1 | 1 | 0% | 1,768 | 2,977 | +68% | 0 | 0 | — |
case-15 | pass→pass | 11,110 | 12,566 | +13% | 1 | 1 | 0% | 2,061 | 2,999 | +46% | 0 | 0 | — |
case-16 | fail→pass | 9,722 | 10,305 | +6% | 1 | 1 | 0% | 1,855 | 2,497 | +35% | 0 | 0 | — |
case-17 | pass→pass | 13,209 | 15,134 | +15% | 1 | 1 | 0% | 2,407 | 3,381 | +40% | 0 | 0 | — |
case-18 | pass→pass | 11,596 | 14,721 | +27% | 1 | 1 | 0% | 2,230 | 3,286 | +47% | 0 | 0 | — |
case-20 | fail→pass | 11,217 | 12,929 | +15% | 1 | 1 | 0% | 2,040 | 2,886 | +41% | 0 | 0 | — |
case-21 | pass→pass | 12,479 | 11,432 | -8% | 1 | 1 | 0% | 2,196 | 2,615 | +19% | 0 | 0 | — |
case-22 | fail→pass | 10,957 | 10,869 | -1% | 1 | 1 | 0% | 2,126 | 2,576 | +21% | 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. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 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.