Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Discover the Spec Kitty 3.2.0 spk skill hierarchy, naming convention, legacy aliases, and the correct skill for a user intent.
.claude/skills/priivacy-ai-spk-meta-skill-map/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -2% | 0% |
Use this skill when the user asks what skills exist, how spk-* names are organized, or which skill applies.
Use spk-<family>-<action-or-topic>.
Legacy family means the pre-3.2.0 spec-kitty-* compatibility skills that remain alongside the newer spk-* hierarchy.
Families:
spk-start-*: onboarding and orientation.spk-mission-*: authoring mission artifacts before runtime execution.spk-run-*: runtime advancement, program orchestration, implementation,review, and blockers.
spk-gate-*: accept, merge, mission review, and retrospectives.spk-admin-*: setup, configuration, upgrades, dashboard/status.spk-team-*: auth, sync, tracker, connectors.spk-doctrine-*: charter, glossary, SPDD, profiles, visual communication,and bulk-edit policy.
spk-integrate-*: APIs, CI, external automation.spk-meta-*: skill discovery and authoring.Use the reference map for the complete inventory: references/spk-skill-map.md.
Choose the narrowest matching skill. If multiple skills match, start at the earliest lifecycle family: start, mission, run, gate, admin/team, doctrine, integrate, meta.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 43,028 | 34,071 | -21% | 1 | 1 | 0% | 2,065 | 988 | -52% | 0 | 0 | — |
case-02 | fail→pass | 19,085 | 8,557 | -55% | 1 | 1 | 0% | 2,595 | 1,654 | -36% | 0 | 0 | — |
case-03 | fail→pass | 9,441 | 38,521 | +308% | 1 | 1 | 0% | 1,579 | 1,767 | +12% | 0 | 0 | — |
case-04 | fail→pass | 13,731 | 4,526 | -67% | 1 | 1 | 0% | 1,794 | 1,182 | -34% | 0 | 0 | — |
case-05 | fail→pass | 8,742 | 4,282 | -51% | 1 | 1 | 0% | 1,123 | 1,103 | -2% | 0 | 0 | — |
case-06 | fail→pass | 7,605 | 3,615 | -52% | 1 | 1 | 0% | 1,116 | 831 | -26% | 0 | 0 | — |
case-07 | pass→pass | 17,269 | 7,246 | -58% | 1 | 1 | 0% | 2,856 | 1,714 | -40% | 0 | 0 | — |
case-08 | fail→pass | 54,388 | 2,772 | -95% | 1 | 1 | 0% | 3,884 | 774 | -80% | 0 | 0 | — |
case-09 | fail→pass | 9,227 | 1,841 | -80% | 1 | 1 | 0% | 1,386 | 596 | -57% | 0 | 0 | — |
case-10 | fail→pass | 12,567 | 2,462 | -80% | 1 | 1 | 0% | 2,016 | 745 | -63% | 0 | 0 | — |
case-11 | fail→pass | 17,235 | 4,324 | -75% | 1 | 1 | 0% | 2,492 | 819 | -67% | 0 | 0 | — |
case-12 | fail→pass | 13,339 | 2,584 | -81% | 1 | 1 | 0% | 1,914 | 753 | -61% | 0 | 0 | — |
case-13 | fail→pass | 22,438 | 3,016 | -87% | 1 | 1 | 0% | 3,642 | 786 | -78% | 0 | 0 | — |
case-14 | fail→pass | 11,415 | 2,365 | -79% | 1 | 1 | 0% | 1,696 | 751 | -56% | 0 | 0 | — |
case-15 | fail→pass | 15,350 | 2,718 | -82% | 1 | 1 | 0% | 2,101 | 789 | -62% | 0 | 0 | — |
case-16 | fail→pass | 12,921 | 3,874 | -70% | 1 | 1 | 0% | 1,758 | 1,128 | -36% | 0 | 0 | — |
case-17 | fail→pass | 26,686 | 2,390 | -91% | 1 | 1 | 0% | 3,437 | 752 | -78% | 0 | 0 | — |
case-18 | fail→pass | 14,058 | 2,405 | -83% | 1 | 1 | 0% | 1,900 | 696 | -63% | 0 | 0 | — |
case-19 | fail→pass | 10,140 | 2,828 | -72% | 1 | 1 | 0% | 1,555 | 774 | -50% | 0 | 0 | — |
case-20 | pass→pass | 8,843 | 3,975 | -55% | 1 | 1 | 0% | 972 | 1,037 | +7% | 0 | 0 | — |
case-21 | pass→pass | 10,621 | 9,926 | -7% | 1 | 1 | 0% | 1,381 | 1,522 | +10% | 0 | 0 | — |
case-22 | pass→pass | 9,441 | 5,415 | -43% | 1 | 1 | 0% | 1,545 | 1,236 | -20% | 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 +82 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/14/2026 | +82% |
Other measured skills in the registry, with their headline benchmark lift.