Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →7-section repo readiness report from `metaharness genome <path>`. Returns repo_type / agent_topology / risk_score / mcp_surface / test_confidence / publish_readiness. Pure-read; degrades gracefully (ADR-150).
.claude/skills/ruvnet-harness-genome/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -23% | 0% |
Companion to harness-score. Where score is a 5-dimension numeric scorecard, genome is a 7-section categorical/numeric report covering repo type, agent topology recommendations, risk score (0-1), MCP surface area, test confidence (0-1), and publish readiness (0-1).
Implementation: scripts/genome.mjs.
npx metaharness genome <path> --json (60s hard timeout).mcp_surface, test_confidence, publish_readiness }.
verdict andverdictExitCode. Upstream exits 1 for needs-work and 2 for blocked; these are valid reports, so the wrapper returns them successfully.
--alert-on-risk-above N: exit 1 when risk_score > N.{
"repo_type": "node_mcp_ci",
"agent_topology": ["maintainer", "tester", "security", "release"],
"risk_score": 0.27,
"mcp_surface": "remote",
"test_confidence": 0.8,
"publish_readiness": 0.9
}Ruflo's risk_score: 0.27 is low (good). publish_readiness: 0.9 is high. The mcp_surface: "remote" reflects that ruflo's MCP servers are hosted, not bundled.
should we?" — genome answers it categorically.
cost-diff-style tooling to spot when agent_topology recommendations drift away from a deliberate architecture choice.
--alert-on-risk-above 0.5 fails the build when the repo'srisk profile crosses a threshold.
harness-score — numeric readinessharness-mcp-scan — static MCP security findingsharness-threat-model — enterprise-review-grade threat model| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 16,561 | 20,129 | +22% | 1 | 1 | 0% | 3,100 | 2,580 | -17% | 0 | 0 | — |
case-01 | fail→pass | 9,805 | 7,895 | -19% | 1 | 1 | 0% | 1,673 | 2,156 | +29% | 0 | 0 | — |
case-02 | fail→fail | 11,978 | 4,676 | -61% | 1 | 1 | 0% | 2,611 | 823 | -68% | 0 | 0 | — |
case-03 | fail→fail | 10,287 | 8,363 | -19% | 1 | 1 | 0% | 1,855 | 2,120 | +14% | 0 | 0 | — |
case-05 | fail→fail | 7,954 | 7,325 | -8% | 1 | 1 | 0% | 709 | 1,370 | +93% | 0 | 0 | — |
case-06 | pass→pass | 25,758 | 22,551 | -12% | 1 | 1 | 0% | 4,539 | 4,585 | +1% | 0 | 0 | — |
case-07 | fail→pass | 18,966 | 4,102 | -78% | 1 | 1 | 0% | 1,953 | 1,522 | -22% | 0 | 0 | — |
case-08 | fail→pass | 12,937 | 5,416 | -58% | 1 | 1 | 0% | 2,133 | 1,678 | -21% | 0 | 0 | — |
case-09 | fail→pass | 10,466 | 5,763 | -45% | 1 | 1 | 0% | 2,019 | 1,628 | -19% | 0 | 0 | — |
case-10 | fail→pass | 7,726 | 3,115 | -60% | 1 | 1 | 0% | 1,525 | 1,179 | -23% | 0 | 0 | — |
case-11 | fail→fail | 14,696 | 3,390 | -77% | 1 | 1 | 0% | 2,892 | 1,317 | -54% | 0 | 0 | — |
case-12 | fail→pass | 14,071 | 4,641 | -67% | 1 | 1 | 0% | 2,705 | 1,518 | -44% | 0 | 0 | — |
case-13 | pass→pass | 7,373 | 3,811 | -48% | 1 | 1 | 0% | 1,225 | 1,231 | +0% | 0 | 0 | — |
case-14 | pass→pass | 6,002 | 2,305 | -62% | 1 | 1 | 0% | 1,073 | 1,046 | -3% | 0 | 0 | — |
case-15 | fail→pass | 14,210 | 2,540 | -82% | 1 | 1 | 0% | 3,075 | 1,042 | -66% | 0 | 0 | — |
case-16 | fail→pass | 11,782 | 945 | -92% | 1 | 1 | 0% | 2,128 | 706 | -67% | 0 | 0 | — |
case-17 | pass→pass | 5,766 | 12,350 | +114% | 1 | 1 | 0% | 926 | 823 | -11% | 0 | 0 | — |
case-18 | pass→pass | 8,807 | 3,899 | -56% | 1 | 1 | 0% | 1,458 | 1,248 | -14% | 0 | 0 | — |
case-19 | fail→pass | 9,712 | 3,209 | -67% | 1 | 1 | 0% | 1,689 | 1,141 | -32% | 0 | 0 | — |
case-20 | fail→pass | 10,021 | 4,754 | -53% | 1 | 1 | 0% | 1,786 | 1,370 | -23% | 0 | 0 | — |
case-21 | fail→pass | 10,671 | 3,893 | -64% | 1 | 1 | 0% | 1,982 | 1,316 | -34% | 0 | 0 | — |
case-22 | pass→pass | 5,966 | 1,754 | -71% | 1 | 1 | 0% | 1,002 | 906 | -10% | 0 | 0 | — |
case-23 | fail→pass | 3,917 | 7,977 | +104% | 1 | 1 | 0% | 745 | 2,185 | +193% | 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. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 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 +52 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.
Other measured skills in the registry, with their headline benchmark lift.