Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyze command history to identify which skills work, which fail, and where to improve.
.claude/skills/sharpdeveye-reflect/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -16% | 0% |
Invoke /agent-workflow — it contains workflow principles, anti-patterns, and the Context Gathering Protocol. Follow the protocol before proceeding — if no workflow context exists yet, you MUST run /teach-maestro first.
Analyze the Maestro audit trail and decision log to produce a skill-effectiveness scorecard. This tells you which commands work, which fail, and where your workflow needs attention.
Read these files from the project root:
.maestro/audit.jsonl — every command invocation with duration, cost, and outcome.maestro/decisions.jsonl — decisions made with outcomes and next stepsIf neither file exists, respond: "No audit data found. Run commands with Maestro to start tracking, then come back."
1. Usage Frequency
2. Completion Rate
3. Command Flow
4. Cost Distribution
5. Duration Analysis
text╔══════════════════════════════════════════╗ ║ MAESTRO EFFECTIVENESS ║ ╠══════════════════════════════════════════╣ ║ Commands Run __ (__ unique) ║ ║ Completion Rate __% ║ ║ Most Used /_____ (__×) ║ ║ Most Abandoned /_____ (__% ⚠️) ║ ║ Avg Duration __s ║ ║ Total Cost ~$__.__ ║ ╠══════════════════════════════════════════╣ ║ STRONGEST PIPELINES ║ ╠══════════════════════════════════════════╣ ║ /_____ → /_____ __× ║ ║ /_____ → /_____ __× ║ ╠══════════════════════════════════════════╣ ║ COST PER COMMAND ║ ╠══════════════════════════════════════════╣ ║ /_____ $__.__/run ████░░ avg ║ ║ /_____ $__.__/run █░░░░░ cheap ║ ║ /_____ $__.__/run █████░ costly ║ ╚══════════════════════════════════════════╝ INSIGHTS: 1. [Data-driven observation with recommended action] 2. [Data-driven observation with recommended action] 3. [Data-driven observation with recommended action]
Every insight MUST:
After reflecting, run /streamline to remove unused commands, or /refine on the most-abandoned command to improve its prompt quality.
NEVER:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | fail→fail | 13,523 | 5,916 | -56% | 1 | 1 | 0% | 2,583 | 1,096 | -58% | 0 | 0 | — |
case-01 | fail→fail | 10,490 | 3,351 | -68% | 1 | 1 | 0% | 1,620 | 1,321 | -18% | 0 | 0 | — |
case-02 | fail→fail | 18,695 | 2,174 | -88% | 1 | 1 | 0% | 3,261 | 1,004 | -69% | 0 | 0 | — |
case-03 | fail→fail | 13,691 | 5,928 | -57% | 1 | 1 | 0% | 2,184 | 1,165 | -47% | 0 | 0 | — |
case-04 | fail→fail | 9,387 | 2,048 | -78% | 1 | 1 | 0% | 1,415 | 1,049 | -26% | 0 | 0 | — |
case-05 | fail→fail | 11,700 | 4,007 | -66% | 1 | 1 | 0% | 1,924 | 1,049 | -45% | 0 | 0 | — |
case-06 | pass→pass | 9,579 | 11,171 | +17% | 1 | 1 | 0% | 1,467 | 2,512 | +71% | 0 | 0 | — |
case-07 | fail→fail | 5,853 | 7,036 | +20% | 1 | 1 | 0% | 147 | 1,299 | +784% | 0 | 0 | — |
case-08 | fail→fail | 2,807 | 6,325 | +125% | 1 | 1 | 0% | 255 | 1,213 | +376% | 0 | 0 | — |
case-10 | pass→fail | 16,909 | 4,840 | -71% | 1 | 1 | 0% | 3,364 | 1,065 | -68% | 0 | 0 | — |
case-11 | fail→pass | 10,660 | 9,980 | -6% | 1 | 1 | 0% | 1,581 | 2,428 | +54% | 0 | 0 | — |
case-12 | fail→fail | 13,410 | 4,741 | -65% | 1 | 1 | 0% | 1,951 | 968 | -50% | 0 | 0 | — |
case-13 | fail→pass | 11,890 | 7,361 | -38% | 1 | 1 | 0% | 1,785 | 1,992 | +12% | 0 | 0 | — |
case-14 | fail→pass | 14,536 | 5,082 | -65% | 1 | 1 | 0% | 2,253 | 1,416 | -37% | 0 | 0 | — |
case-15 | pass→pass | 11,117 | 3,288 | -70% | 1 | 1 | 0% | 1,632 | 1,333 | -18% | 0 | 0 | — |
case-16 | fail→fail | 10,722 | 2,444 | -77% | 1 | 1 | 0% | 1,533 | 1,106 | -28% | 0 | 0 | — |
case-17 | fail→pass | 12,269 | 11,662 | -5% | 1 | 1 | 0% | 1,888 | 2,715 | +44% | 0 | 0 | — |
case-18 | fail→fail | 14,755 | 5,145 | -65% | 1 | 1 | 0% | 2,374 | 994 | -58% | 0 | 0 | — |
case-19 | fail→fail | 17,707 | 9,691 | -45% | 1 | 1 | 0% | 1,632 | 2,078 | +27% | 0 | 0 | — |
case-20 | fail→fail | 12,365 | 7,832 | -37% | 1 | 1 | 0% | 2,386 | 1,289 | -46% | 0 | 0 | — |
case-21 | fail→fail | 4,267 | 6,759 | +58% | 1 | 1 | 0% | 560 | 1,402 | +150% | 0 | 0 | — |
case-22 | fail→pass | 16,003 | 5,955 | -63% | 1 | 1 | 0% | 2,208 | 1,856 | -16% | 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 12 counted toward the lift figure. The other 10 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 +18 percentage points is the difference between those two pass rates over the 12 comparable cases. 2 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.