Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Runs plugin health checks (venv packages, skill registration, and album slug collisions). Use when the user asks to check plugin health, verify setup, or troubleshoot missing skills.
.claude/skills/bitwize-music-studio-health-check/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-19 | ✗→✓ | ▲ Improved | -72% | 0% |
Run the health_check MCP tool and report results to the user.
IMPORTANT: Do NOT use Bash for any step. Use only the tools listed below.
ToolSearch tool with query select:mcp__plugin_bitwize-music_bitwize-music-mcp__health_check to load the MCP tool schemamcp__plugin_bitwize-music_bitwize-music-mcp__health_check (the MCP tool, not a CLI command)HEALTH CHECK: OK
Venv: N packages verified
Skills: N skills registered
Collisions: no album slug collisionsHEALTH CHECK: WARN
VENV [warn]
N outdated: pkg1 (1.0 -> 1.1), pkg2 (2.0 -> 2.1)
N missing: pkg3, pkg4
Fix: <the venv check's `fix` field from the health_check result> (already the correct command for the user's OS)
SKILLS [warn]
N missing from Claude Code: skill-a, skill-b
N ghost (deleted but cached): skill-c
Fix: claude plugin update bitwize-music
COLLISIONS [warn]
N album slug collision(s):
slug-name: kept [genre-a], shadowed by [genre-b]
Fix: Rename one album with /bitwize-music:rename or move its directory, then run rebuild_state
For comprehensive diagnostics, run the `diagnose` MCP tool.HEALTH CHECK: FAIL
VENV [fail]
Venv not found at ~/.bitwize-music/venv
Fix: /bitwize-music:setupdiagnose MCP tool for deeper checks| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 7,944 | 4,715 | -41% | 1 | 1 | 0% | 1,405 | 696 | -50% | 0 | 0 | — |
case-02 | fail→fail | 10,570 | 4,510 | -57% | 1 | 1 | 0% | 1,778 | 710 | -60% | 0 | 0 | — |
case-03 | fail→fail | 5,933 | 4,327 | -27% | 1 | 1 | 0% | 1,025 | 691 | -33% | 0 | 0 | — |
case-04 | fail→fail | 4,313 | 4,761 | +10% | 1 | 1 | 0% | 733 | 755 | +3% | 0 | 0 | — |
case-05 | fail→fail | 10,328 | 4,445 | -57% | 1 | 1 | 0% | 1,536 | 674 | -56% | 0 | 0 | — |
case-06 | fail→fail | 9,610 | 4,955 | -48% | 1 | 1 | 0% | 1,841 | 915 | -50% | 0 | 0 | — |
case-07 | fail→pass | 8,698 | 2,399 | -72% | 1 | 1 | 0% | 1,530 | 912 | -40% | 0 | 0 | — |
case-08 | fail→fail | 8,947 | 6,794 | -24% | 1 | 1 | 0% | 1,507 | 795 | -47% | 0 | 0 | — |
case-09 | fail→fail | 8,257 | 5,768 | -30% | 1 | 1 | 0% | 1,387 | 1,021 | -26% | 0 | 0 | — |
case-10 | fail→pass | 8,751 | 1,722 | -80% | 1 | 1 | 0% | 1,453 | 746 | -49% | 0 | 0 | — |
case-11 | fail→pass | 8,115 | 2,375 | -71% | 1 | 1 | 0% | 1,346 | 895 | -34% | 0 | 0 | — |
case-12 | fail→fail | 9,467 | 2,012 | -79% | 1 | 1 | 0% | 1,553 | 849 | -45% | 0 | 0 | — |
case-13 | fail→pass | 7,357 | 1,961 | -73% | 1 | 1 | 0% | 1,191 | 869 | -27% | 0 | 0 | — |
case-14 | pass→fail | 7,987 | 3,166 | -60% | 1 | 1 | 0% | 1,297 | 970 | -25% | 0 | 0 | — |
case-15 | fail→fail | 5,304 | 6,177 | +16% | 1 | 1 | 0% | 773 | 928 | +20% | 0 | 0 | — |
case-16 | pass→pass | 7,401 | 3,055 | -59% | 1 | 1 | 0% | 1,458 | 987 | -32% | 0 | 0 | — |
case-17 | pass→pass | 6,028 | 2,768 | -54% | 1 | 1 | 0% | 1,009 | 883 | -12% | 0 | 0 | — |
case-18 | pass→pass | 7,207 | 5,117 | -29% | 1 | 1 | 0% | 1,165 | 1,328 | +14% | 0 | 0 | — |
case-19 | fail→pass | 14,850 | 1,794 | -88% | 1 | 1 | 0% | 2,548 | 726 | -72% | 0 | 0 | — |
case-20 | pass→fail | 12,288 | 11,743 | -4% | 1 | 1 | 0% | 1,966 | 1,540 | -22% | 0 | 0 | — |
case-21 | pass→fail | 12,187 | 12,238 | +0% | 1 | 1 | 0% | 2,310 | 1,753 | -24% | 0 | 0 | — |
case-22 | pass→pass | 17,442 | 11,067 | -37% | 1 | 1 | 0% | 2,748 | 2,460 | -10% | 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, and 11 counted toward the lift figure. The other 11 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +9 percentage points is the difference between those two pass rates over the 11 comparable cases. 4 cases got worse with the skill loaded, and they are 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.