Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Full job search coaching system — JD decoding, resume, storybank, mock interviews, transcript analysis, comp negotiation. 23 commands, persistent state.
.claude/skills/sickn33-interview-coach/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -56% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -61% | 0% |
A persistent, adaptive coaching system for the full job search lifecycle. Not a question bank — an opinionated system that tracks your patterns, scores your answers, and gets sharper the more you use it. State persists in coaching_state.md across sessions so you always pick up where you left off.
bashnpx skills add dbhat93/job-search-os
Then type /coach → kickoff.
/coach
kickoffThe coach asks for your resume, target role, and timeline — then builds your profile and gives you a prioritized action plan.
/coach
prep Stripe Senior PMRuns company research, generates a role-specific prep brief, and queues up mock interview questions tailored to Stripe's process.
/coach
analyzePaste a raw transcript from Otter, Zoom, or any tool. The coach auto-detects the format, scores each answer across five dimensions, and gives you a drill plan targeting your specific gaps.
/coach
salaryCoaches you through the recruiter screen "what are your salary expectations?" moment with a defensible range and exact scripts.
https://github.com/dbhat93/job-search-os
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | pass→pass | 18,671 | 10,238 | -45% | 1 | 1 | 0% | 2,726 | 2,247 | -18% | 0 | 0 | — |
case-14 | pass→pass | 17,682 | 7,702 | -56% | 1 | 1 | 0% | 2,723 | 1,692 | -38% | 0 | 0 | — |
case-19 | fail→fail | 16,484 | 13,732 | -17% | 1 | 1 | 0% | 2,315 | 2,552 | +10% | 0 | 0 | — |
case-01 | fail→fail | 6,070 | 3,799 | -37% | 1 | 1 | 0% | 958 | 1,187 | +24% | 0 | 0 | — |
case-02 | fail→pass | 6,882 | 2,158 | -69% | 1 | 1 | 0% | 922 | 950 | +3% | 0 | 0 | — |
case-03 | fail→pass | 11,523 | 1,834 | -84% | 1 | 1 | 0% | 1,796 | 788 | -56% | 0 | 0 | — |
case-04 | fail→pass | 10,617 | 2,980 | -72% | 1 | 1 | 0% | 1,561 | 1,040 | -33% | 0 | 0 | — |
case-05 | fail→pass | 9,892 | 1,846 | -81% | 1 | 1 | 0% | 1,520 | 865 | -43% | 0 | 0 | — |
case-06 | pass→fail | 13,213 | 2,365 | -82% | 1 | 1 | 0% | 1,860 | 927 | -50% | 0 | 0 | — |
case-07 | fail→pass | 18,066 | 2,350 | -87% | 1 | 1 | 0% | 2,414 | 949 | -61% | 0 | 0 | — |
case-08 | fail→fail | 10,070 | 2,203 | -78% | 1 | 1 | 0% | 1,558 | 905 | -42% | 0 | 0 | — |
case-09 | fail→pass | 14,114 | 4,469 | -68% | 1 | 1 | 0% | 1,952 | 1,264 | -35% | 0 | 0 | — |
case-10 | pass→pass | 9,591 | 2,068 | -78% | 1 | 1 | 0% | 1,364 | 897 | -34% | 0 | 0 | — |
case-11 | fail→pass | 5,784 | 1,385 | -76% | 1 | 1 | 0% | 795 | 751 | -6% | 0 | 0 | — |
case-12 | fail→pass | 5,646 | 1,601 | -72% | 1 | 1 | 0% | 772 | 781 | +1% | 0 | 0 | — |
case-15 | pass→pass | 8,482 | 6,777 | -20% | 1 | 1 | 0% | 1,357 | 1,630 | +20% | 0 | 0 | — |
case-16 | fail→pass | 13,852 | 1,622 | -88% | 1 | 1 | 0% | 1,981 | 825 | -58% | 0 | 0 | — |
case-17 | pass→pass | 14,443 | 3,075 | -79% | 1 | 1 | 0% | 2,160 | 1,053 | -51% | 0 | 0 | — |
case-18 | pass→pass | 8,033 | 1,999 | -75% | 1 | 1 | 0% | 1,175 | 879 | -25% | 0 | 0 | — |
case-20 | fail→pass | 10,452 | 11,925 | +14% | 1 | 1 | 0% | 1,460 | 2,292 | +57% | 0 | 0 | — |
case-21 | pass→pass | 4,462 | 4,821 | +8% | 1 | 1 | 0% | 613 | 1,273 | +108% | 0 | 0 | — |
case-22 | pass→pass | 4,001 | 5,548 | +39% | 1 | 1 | 0% | 535 | 1,214 | +127% | 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 +41 percentage points is the difference between those two pass rates over the 22 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.