Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →5-dimension harness readiness scorecard from `metaharness score <path>`. Returns harnessFit / compileConfidence / taskCoverage / toolSafety / memoryUsefulness + estCostPerRunUsd + scaffoldReady. Pure-read; subprocess invocation; degrades gracefully when MetaHarness is absent (ADR-150 architectural constraint).
.claude/skills/ruvnet-harness-score/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -47% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -37% | 0% |
Surfaces the upstream metaharness score CLI as a ruflo skill. Use when Claude Code needs to assess whether a repo is ready for harness adoption before recommending the user run npx ruflo init or harness-mint.
Implementation: scripts/score.mjs.
npx metaharness score <path> --json (single subprocess,60s hard timeout).
toolSafety, memoryUsefulness, estCostPerRunUsd, recommendedMode, archetype, template, scaffoldReady, hardConstraints }.
--alert-on-fit-below N: exit 1 when harnessFit < N.| Dimension | Value | |---|---:| | harnessFit | 82/100 | | compileConfidence | 100 | | taskCoverage | 79 | | toolSafety | 100 | | memoryUsefulness | 40 | | estCostPerRunUsd | $0.048 | | recommendedMode | CLI + MCP | | archetype | typescript-sdk-harness | | template | vertical:coding | | scaffoldReady | true |
Ruflo passes its own readiness check. memoryUsefulness: 40 is the weakest dimension — track this as a leading indicator for future memory work in the AgentDB layer.
bashnode plugins/ruflo-metaharness/scripts/score.mjs --alert-on-fit-below 70 --format json
Exit 1 fails the build. Pair with harness-genome for the full 7-section view.
When metaharness is not installed and npx can't fetch it (offline, no network, registry unreachable), the script emits:
json{ "degraded": true, "reason": "metaharness-not-available", "hint": "Install with `npm i -D metaharness@~0.3.0` (pinned range — this plugin never fetches @latest) or verify network access for the one-time cache install." }
and exits 0. Ruflo continues to function — this is the architectural constraint in action.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→pass | 18,206 | 1,456 | -92% | 1 | 1 | 0% | 1,030 | 859 | -17% | 0 | 0 | — |
case-01 | fail→fail | 2,167 | 2,910 | +34% | 1 | 1 | 0% | 254 | 930 | +266% | 0 | 0 | — |
case-02 | pass→fail | 13,024 | 17,638 | +35% | 1 | 1 | 0% | 2,056 | 837 | -59% | 0 | 0 | — |
case-03 | fail→pass | 13,560 | 4,328 | -68% | 1 | 1 | 0% | 2,797 | 1,495 | -47% | 0 | 0 | — |
case-04 | fail→pass | 8,853 | 2,249 | -75% | 1 | 1 | 0% | 1,599 | 967 | -40% | 0 | 0 | — |
case-05 | fail→pass | 11,842 | 2,902 | -75% | 1 | 1 | 0% | 2,173 | 1,110 | -49% | 0 | 0 | — |
case-06 | fail→pass | 7,873 | 1,793 | -77% | 1 | 1 | 0% | 1,438 | 905 | -37% | 0 | 0 | — |
case-08 | pass→pass | 5,052 | 1,843 | -64% | 1 | 1 | 0% | 884 | 877 | -1% | 0 | 0 | — |
case-09 | fail→pass | 7,870 | 1,930 | -75% | 1 | 1 | 0% | 1,282 | 911 | -29% | 0 | 0 | — |
case-10 | pass→pass | 4,491 | 1,510 | -66% | 1 | 1 | 0% | 741 | 869 | +17% | 0 | 0 | — |
case-11 | fail→pass | 8,860 | 1,695 | -81% | 1 | 1 | 0% | 1,569 | 889 | -43% | 0 | 0 | — |
case-12 | fail→pass | 8,582 | 1,426 | -83% | 1 | 1 | 0% | 1,634 | 770 | -53% | 0 | 0 | — |
case-13 | pass→pass | 9,187 | 4,424 | -52% | 1 | 1 | 0% | 1,541 | 928 | -40% | 0 | 0 | — |
case-14 | pass→pass | 7,025 | 1,628 | -77% | 1 | 1 | 0% | 1,348 | 865 | -36% | 0 | 0 | — |
case-15 | fail→pass | 7,907 | 3,087 | -61% | 1 | 1 | 0% | 1,514 | 865 | -43% | 0 | 0 | — |
case-16 | fail→pass | 5,460 | 1,865 | -66% | 1 | 1 | 0% | 808 | 865 | +7% | 0 | 0 | — |
case-17 | fail→pass | 6,544 | 2,335 | -64% | 1 | 1 | 0% | 1,063 | 959 | -10% | 0 | 0 | — |
case-18 | fail→pass | 9,975 | 1,720 | -83% | 1 | 1 | 0% | 1,850 | 854 | -54% | 0 | 0 | — |
case-19 | fail→pass | 12,569 | 1,589 | -87% | 1 | 1 | 0% | 2,157 | 887 | -59% | 0 | 0 | — |
case-20 | fail→pass | 6,508 | 2,195 | -66% | 1 | 1 | 0% | 1,031 | 984 | -5% | 0 | 0 | — |
case-21 | pass→pass | 7,437 | 4,042 | -46% | 1 | 1 | 0% | 1,315 | 1,399 | +6% | 0 | 0 | — |
case-22 | fail→pass | 4,027 | 2,376 | -41% | 1 | 1 | 0% | 736 | 986 | +34% | 0 | 0 | — |
case-23 | fail→pass | 13,474 | 7,818 | -42% | 1 | 1 | 0% | 2,474 | 2,040 | -18% | 0 | 0 | — |
case-24 | pass→pass | 16,699 | 6,063 | -64% | 1 | 1 | 0% | 3,643 | 1,673 | -54% | 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. 24 cases were attempted, and 22 counted toward the lift figure. The other 2 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 +63 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.