Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Tri-model orchestration — run the same task through Claude + Codex + Gemini in parallel and synthesize the best answer. Use for high-stakes decisions where multi-model consensus reduces single-model bias.
.claude/skills/evolution-foundation-dev-ccg/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 0% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -24% | 0% |
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
EXPERIMENTAL. Tri-model orchestration: run the same prompt through Claude, Codex (OpenAI Codex / GPT-Code), and Gemini in parallel, then synthesize the best answer.
> Note: This skill assumes external API access to Codex and Gemini, which may or may not be configured in EvoNexus. If only Claude is available, the skill degrades to single-model mode and warns the user.
workspace/development/research/[C]ccg-{topic}-{date}.mdmarkdown## CCG Synthesis — {topic} ### Models Consulted - ✅ Claude (Opus 4.6) - ✅ Codex (gpt-code-X) - ✅ Gemini (gemini-Y) ### Agreement Zones - All three agree on: {points} ### Disagreement Zones - Claude says X - Codex says Y - Gemini says Z - → Flagged for human review ### Synthesized Answer [Combined reasoning, marking uncertainty zones]
@apex-architect (for high-stakes architectural decisions)@raven-critic (for adversarial cross-validation)If Codex/Gemini APIs are not configured, this skill warns: > "CCG requires Codex and Gemini API access. Currently only Claude is available — falling back to single-model mode. Configure via dev-mcp-setup if needed."
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | pass→pass | 14,161 | 5,693 | -60% | 1 | 1 | 0% | 2,116 | 1,315 | -38% | 0 | 0 | — |
case-01 | fail→pass | 32,831 | 22,329 | -32% | 1 | 1 | 0% | 5,288 | 4,139 | -22% | 0 | 0 | — |
case-02 | fail→pass | 24,278 | 20,193 | -17% | 1 | 1 | 0% | 3,912 | 3,896 | -0% | 0 | 0 | — |
case-03 | fail→pass | 38,787 | 27,173 | -30% | 1 | 1 | 0% | 6,203 | 4,740 | -24% | 0 | 0 | — |
case-04 | pass→pass | 2,396 | 2,999 | +25% | 1 | 1 | 0% | 442 | 1,043 | +136% | 0 | 0 | — |
case-05 | pass→pass | 2,392 | 2,786 | +16% | 1 | 1 | 0% | 358 | 966 | +170% | 0 | 0 | — |
case-06 | pass→pass | 3,155 | 2,320 | -26% | 1 | 1 | 0% | 515 | 887 | +72% | 0 | 0 | — |
case-07 | fail→pass | 14,660 | 7,571 | -48% | 1 | 1 | 0% | 2,161 | 1,858 | -14% | 0 | 0 | — |
case-08 | fail→pass | 7,368 | 2,644 | -64% | 1 | 1 | 0% | 1,361 | 1,028 | -24% | 0 | 0 | — |
case-10 | fail→pass | 12,114 | 8,335 | -31% | 1 | 1 | 0% | 1,967 | 1,899 | -3% | 0 | 0 | — |
case-11 | pass→pass | 7,625 | 2,454 | -68% | 1 | 1 | 0% | 1,159 | 953 | -18% | 0 | 0 | — |
case-12 | fail→pass | 26,247 | 18,322 | -30% | 1 | 1 | 0% | 4,041 | 3,344 | -17% | 0 | 0 | — |
case-13 | fail→pass | 23,577 | 18,772 | -20% | 1 | 1 | 0% | 3,563 | 3,474 | -2% | 0 | 0 | — |
case-14 | fail→pass | 21,969 | 17,266 | -21% | 1 | 1 | 0% | 3,183 | 3,162 | -1% | 0 | 0 | — |
case-15 | pass→pass | 8,994 | 7,060 | -22% | 1 | 1 | 0% | 1,354 | 1,658 | +22% | 0 | 0 | — |
case-16 | pass→pass | 5,424 | 7,798 | +44% | 1 | 1 | 0% | 817 | 1,702 | +108% | 0 | 0 | — |
case-17 | fail→pass | 8,649 | 2,314 | -73% | 1 | 1 | 0% | 1,478 | 878 | -41% | 0 | 0 | — |
case-18 | fail→fail | 17,245 | 17,466 | +1% | 1 | 1 | 0% | 2,638 | 3,365 | +28% | 0 | 0 | — |
case-19 | pass→pass | 13,256 | 11,868 | -10% | 1 | 1 | 0% | 2,065 | 2,442 | +18% | 0 | 0 | — |
case-20 | pass→pass | 13,577 | 5,342 | -61% | 1 | 1 | 0% | 2,095 | 1,365 | -35% | 0 | 0 | — |
case-21 | fail→pass | 12,163 | 2,762 | -77% | 1 | 1 | 0% | 1,830 | 980 | -46% | 0 | 0 | — |
case-22 | fail→pass | 12,554 | 1,385 | -89% | 1 | 1 | 0% | 1,919 | 699 | -64% | 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 +55 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.