Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deep codebase initialization — generate hierarchical AGENTS.md / CLAUDE.md context files for a new project so engineering agents have full context from session start.
.claude/skills/evolution-foundation-dev-deepinit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -34% | 0% |
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
Deep codebase initialization. For a new project, generate hierarchical AGENTS.md / CLAUDE.md context files that give engineering agents full project context from session start — instead of forcing them to re-discover the codebase every session.
Delegate to @scout-explorer:
Create the hierarchical context:
{project-root}/
├── CLAUDE.md # Top-level project context (this is what Claude reads first)
├── AGENTS.md # Detailed context for engineering agents
└── docs/
└── architecture/
└── [C]project-overview-{date}.md@oath-verifier reads the new context docs and confirms they're accuratemarkdown# {Project Name} [1-2 sentence purpose] ## Tech Stack - Language: ... - Framework: ... - Build: ... ## Quick Commands - Build: `...` - Test: `...` - Run: `...` ## Architecture [high-level overview] ## Conventions [detected patterns]
markdown# Engineering Context ## Module Map - `src/auth/` — authentication - `src/db/` — database layer - ... ## Key Files - `src/main.ts:42` — entry point - ... ## Patterns [deep convention notes] ## Gotchas [known pitfalls]
@scout-explorer (Phase 1)@quill-writer (writes the actual docs)@oath-verifier (Phase 4 verification)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,281 | 2,188 | -87% | 1 | 1 | 0% | 2,764 | 971 | -65% | 0 | 0 | — |
case-02 | fail→fail | 3,226 | 2,558 | -21% | 1 | 1 | 0% | 239 | 977 | +309% | 0 | 0 | — |
case-03 | fail→fail | 17,099 | 2,700 | -84% | 1 | 1 | 0% | 2,965 | 1,116 | -62% | 0 | 0 | — |
case-04 | fail→pass | 7,919 | 1,681 | -79% | 1 | 1 | 0% | 1,247 | 841 | -33% | 0 | 0 | — |
case-05 | fail→pass | 8,893 | 1,798 | -80% | 1 | 1 | 0% | 1,424 | 915 | -36% | 0 | 0 | — |
case-06 | pass→pass | 8,041 | 2,067 | -74% | 1 | 1 | 0% | 1,418 | 972 | -31% | 0 | 0 | — |
case-07 | fail→pass | 8,172 | 1,594 | -80% | 1 | 1 | 0% | 1,251 | 849 | -32% | 0 | 0 | — |
case-08 | pass→pass | 2,395 | 2,077 | -13% | 1 | 1 | 0% | 370 | 901 | +144% | 0 | 0 | — |
case-09 | pass→pass | 6,700 | 1,704 | -75% | 1 | 1 | 0% | 1,041 | 872 | -16% | 0 | 0 | — |
case-10 | fail→pass | 9,406 | 2,454 | -74% | 1 | 1 | 0% | 1,537 | 1,021 | -34% | 0 | 0 | — |
case-11 | pass→pass | 13,262 | 7,887 | -41% | 1 | 1 | 0% | 2,249 | 2,011 | -11% | 0 | 0 | — |
case-12 | fail→pass | 12,411 | 4,576 | -63% | 1 | 1 | 0% | 2,026 | 1,332 | -34% | 0 | 0 | — |
case-13 | pass→pass | 13,115 | 3,641 | -72% | 1 | 1 | 0% | 1,927 | 1,206 | -37% | 0 | 0 | — |
case-14 | pass→pass | 8,152 | 3,225 | -60% | 1 | 1 | 0% | 1,268 | 1,143 | -10% | 0 | 0 | — |
case-15 | pass→pass | 8,554 | 3,308 | -61% | 1 | 1 | 0% | 1,336 | 1,181 | -12% | 0 | 0 | — |
case-20 | pass→pass | 6,613 | 2,173 | -67% | 1 | 1 | 0% | 973 | 933 | -4% | 0 | 0 | — |
case-16 | pass→pass | 12,689 | 8,531 | -33% | 1 | 1 | 0% | 2,065 | 2,106 | +2% | 0 | 0 | — |
case-17 | fail→pass | 12,625 | 4,639 | -63% | 1 | 1 | 0% | 1,888 | 1,485 | -21% | 0 | 0 | — |
case-18 | fail→fail | 8,442 | 2,588 | -69% | 1 | 1 | 0% | 1,262 | 967 | -23% | 0 | 0 | — |
case-19 | pass→pass | 14,637 | 9,316 | -36% | 1 | 1 | 0% | 2,414 | 2,273 | -6% | 0 | 0 | — |
case-21 | pass→pass | 14,062 | 4,444 | -68% | 1 | 1 | 0% | 2,278 | 1,372 | -40% | 0 | 0 | — |
case-22 | fail→pass | 4,876 | 2,252 | -54% | 1 | 1 | 0% | 602 | 1,015 | +69% | 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 21 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 +32 percentage points is the difference between those two pass rates over the 21 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.