Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Dispatch docs subagent to audit, generate, and validate documentation against the codebase.
.claude/skills/automagik-dev-docs/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 2% | 0% |
Runtime syntax: invoke the plugin copy through the active runtime's owner-qualified skill selector; use a bare selector only when intentionally selecting a user-tier copy (a separately installed personal copy; Genie no longer seeds this tier). Cross-skill prose below uses bare names as portable semantic routes; the orchestrator resolves the selector for the active runtime.
Audit existing documentation, fill gaps, and validate every claim against actual code. Standalone or as part of work.
work completes)| Type | Location | Purpose | |------|----------|---------| | README | README.md, */README.md | Overview, setup, usage | | AGENTS.md | AGENTS.md, */AGENTS.md | Agent conventions, constraints, commands, verification | | CLAUDE.md | CLAUDE.md, */CLAUDE.md | Conventions, commands, gotchas for agents | | API docs | docs/api/, inline JSDoc/TSDoc | Contracts, request/response schemas | | Architecture | docs/architecture.md, ARCHITECTURE.md | System design, data flow | | Inline | JSDoc, TSDoc, docstrings | Function/class/module docs |
AGENTS.md is the governing agent instruction surface. CLAUDE.md remains evidence of repository intent when present; keep both current when the project has both files.
AGENTS.md accuracy first.Runs as a subagent (native runtime): the dispatching agent issues an native delegation surface call with a curated brief — scope (which docs, which change triggered the audit), the code areas to validate against, and the expected report shape.
Example brief: "Audit README.md, CLAUDE.md, and skills/work/SKILL.md after PR #746 — verify dispatch examples match current code, fix stale references, report per-file verdicts with evidence."
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | pass→pass | 12,257 | 5,446 | -56% | 1 | 1 | 0% | 1,972 | 1,733 | -12% | 0 | 0 | — |
case-04 | fail→fail | 12,108 | 9,111 | -25% | 1 | 1 | 0% | 2,004 | 1,012 | -50% | 0 | 0 | — |
case-05 | pass→fail | 5,203 | 4,990 | -4% | 1 | 1 | 0% | 911 | 872 | -4% | 0 | 0 | — |
case-01 | fail→fail | 16,034 | 5,618 | -65% | 1 | 1 | 0% | 2,957 | 992 | -66% | 0 | 0 | — |
case-02 | fail→fail | 3,956 | 5,961 | +51% | 1 | 1 | 0% | 188 | 1,089 | +479% | 0 | 0 | — |
case-03 | pass→pass | 8,956 | 18,744 | +109% | 1 | 1 | 0% | 1,809 | 3,227 | +78% | 0 | 0 | — |
case-07 | fail→pass | 16,839 | 10,856 | -36% | 1 | 1 | 0% | 2,506 | 2,348 | -6% | 0 | 0 | — |
case-08 | fail→pass | 12,499 | 6,441 | -48% | 1 | 1 | 0% | 1,812 | 1,755 | -3% | 0 | 0 | — |
case-09 | fail→pass | 12,035 | 6,568 | -45% | 1 | 1 | 0% | 1,770 | 1,700 | -4% | 0 | 0 | — |
case-10 | pass→pass | 14,090 | 9,341 | -34% | 1 | 1 | 0% | 2,140 | 2,197 | +3% | 0 | 0 | — |
case-11 | pass→pass | 11,470 | 7,051 | -39% | 1 | 1 | 0% | 1,893 | 1,890 | -0% | 0 | 0 | — |
case-12 | pass→pass | 7,017 | 2,857 | -59% | 1 | 1 | 0% | 1,153 | 1,165 | +1% | 0 | 0 | — |
case-13 | pass→pass | 12,299 | 5,201 | -58% | 1 | 1 | 0% | 2,036 | 1,574 | -23% | 0 | 0 | — |
case-14 | pass→pass | 13,613 | 7,555 | -45% | 1 | 1 | 0% | 2,042 | 1,872 | -8% | 0 | 0 | — |
case-15 | fail→pass | 12,436 | 5,120 | -59% | 1 | 1 | 0% | 2,002 | 1,480 | -26% | 0 | 0 | — |
case-16 | fail→pass | 8,535 | 3,887 | -54% | 1 | 1 | 0% | 1,322 | 1,347 | +2% | 0 | 0 | — |
case-17 | pass→pass | 13,070 | 7,185 | -45% | 1 | 1 | 0% | 1,982 | 1,758 | -11% | 0 | 0 | — |
case-18 | fail→pass | 12,600 | 3,593 | -71% | 1 | 1 | 0% | 1,990 | 1,332 | -33% | 0 | 0 | — |
case-19 | pass→pass | 12,452 | 6,452 | -48% | 1 | 1 | 0% | 1,833 | 1,662 | -9% | 0 | 0 | — |
case-20 | fail→pass | 12,444 | 6,828 | -45% | 1 | 1 | 0% | 1,941 | 1,909 | -2% | 0 | 0 | — |
case-21 | pass→pass | 13,159 | 2,381 | -82% | 1 | 1 | 0% | 1,970 | 1,072 | -46% | 0 | 0 | — |
case-22 | fail→pass | 10,708 | 2,177 | -80% | 1 | 1 | 0% | 1,553 | 1,029 | -34% | 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 18 counted toward the lift figure. The other 4 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 18 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.