Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Triage review findings interactively — approve, skip, or prioritize each issue. Use after phx-review to filter findings before fixing.
.claude/skills/oliver-kriska-phx-triage/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -22% | 0% |
Walk through review findings one by one for human decision before committing to fixes.
/skill:phx-triage .claude/plans/user-auth/reviews/user-auth-review.md
/skill:phx-triage # Uses most recent reviewAfter /skill:phx-review produces findings, you have three options:
/skill:phx-plan .claude/plans/{slug}/reviews/.../skill:phx-triage (filter, then fix what matters)Best when review has 5+ findings and you want to prioritize.
Read the review file. Parse all findings with severity.
Auto-approve Iron Law violations: Findings matching any Iron Law are auto-approved as "Fix it" without asking. These are non-negotiable in Elixir/Phoenix development.
Use AskUserQuestion with multiSelect: true. Start with severity shortcuts, then list individual findings:
AskUserQuestion:
question: "Which findings do you want to fix? (Iron Law violations auto-included)"
header: "Triage"
multiSelect: true
options:
- label: "All BLOCKERs ({count})"
description: "Fix all critical issues"
- label: "All WARNINGs ({count})"
description: "Fix all should-fix issues"
- label: "[BLOCKER] {title 1}"
description: "{file}:{line} — {brief description}"
- label: "[WARNING] {title 2}"
description: "{file}:{line} — {brief description}"If >4 options, batch into groups of 4 with severity shortcuts in the first batch. Severity shortcuts select all findings of that level — user can mix shortcuts with individual picks.
AskUserQuestion discipline: decisions only — never use it to narrate progress. Every option carries {file}:{line} + counts so the user decides without opening the review file.
For selected items, ask ONE batch follow-up: "Any specific approach for any of these?" If they say "just fix them", proceed.
Write to .claude/plans/{slug}/reviews/{slug}-triage.md with Fix Queue (approved items with checkboxes), Skipped, and Deferred sections.
Triage complete: {n} to fix, {n} skipped, {n} deferred.
1. Plan fixes — /skill:phx-plan .claude/plans/{slug}/reviews/{slug}-triage.md
2. Fix directly — /skill:phx-work (for simple fixes)
3. Review deferred items later
4. Capture solutions — /skill:phx-compound (if patterns were solved)any user guidance for better fixes
(root cause identified, fix known), mention /skill:phx-compound
text/skill:phx-review | /skill:phx-triage ← YOU ARE HERE (interactive filtering) | /skill:phx-plan (with triage file) → /skill:phx-work → /skill:phx-compound
references/triage-patterns.md — Common triage decisions| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | pass→pass | 12,483 | 4,128 | -67% | 1 | 1 | 0% | 1,792 | 1,543 | -14% | 0 | 0 | — |
case-01 | fail→fail | 5,763 | 2,893 | -50% | 1 | 1 | 0% | 913 | 1,236 | +35% | 0 | 0 | — |
case-02 | fail→fail | 5,793 | 5,648 | -3% | 1 | 1 | 0% | 305 | 1,259 | +313% | 0 | 0 | — |
case-03 | fail→fail | 7,390 | 4,422 | -40% | 1 | 1 | 0% | 1,130 | 1,142 | +1% | 0 | 0 | — |
case-04 | fail→pass | 7,098 | 2,057 | -71% | 1 | 1 | 0% | 1,057 | 1,278 | +21% | 0 | 0 | — |
case-05 | fail→pass | 14,002 | 6,253 | -55% | 1 | 1 | 0% | 2,178 | 2,059 | -5% | 0 | 0 | — |
case-06 | fail→fail | 11,135 | 3,288 | -70% | 1 | 1 | 0% | 1,564 | 1,476 | -6% | 0 | 0 | — |
case-07 | fail→pass | 8,644 | 2,514 | -71% | 1 | 1 | 0% | 1,249 | 1,345 | +8% | 0 | 0 | — |
case-08 | pass→pass | 12,942 | 4,921 | -62% | 1 | 1 | 0% | 1,883 | 1,759 | -7% | 0 | 0 | — |
case-10 | fail→pass | 9,118 | 3,166 | -65% | 1 | 1 | 0% | 1,308 | 1,492 | +14% | 0 | 0 | — |
case-11 | fail→pass | 10,316 | 2,160 | -79% | 1 | 1 | 0% | 1,626 | 1,269 | -22% | 0 | 0 | — |
case-12 | fail→pass | 7,714 | 2,263 | -71% | 1 | 1 | 0% | 1,097 | 1,287 | +17% | 0 | 0 | — |
case-13 | fail→pass | 11,109 | 2,600 | -77% | 1 | 1 | 0% | 1,674 | 1,321 | -21% | 0 | 0 | — |
case-14 | fail→pass | 12,253 | 2,986 | -76% | 1 | 1 | 0% | 1,695 | 1,456 | -14% | 0 | 0 | — |
case-15 | fail→pass | 9,789 | 2,018 | -79% | 1 | 1 | 0% | 1,417 | 1,223 | -14% | 0 | 0 | — |
case-16 | pass→pass | 8,915 | 3,100 | -65% | 1 | 1 | 0% | 1,304 | 1,416 | +9% | 0 | 0 | — |
case-17 | pass→pass | 11,011 | 3,681 | -67% | 1 | 1 | 0% | 1,714 | 1,504 | -12% | 0 | 0 | — |
case-18 | fail→pass | 9,674 | 3,565 | -63% | 1 | 1 | 0% | 1,586 | 1,531 | -3% | 0 | 0 | — |
case-19 | pass→fail | 18,063 | 5,422 | -70% | 1 | 1 | 0% | 3,621 | 1,316 | -64% | 0 | 0 | — |
case-20 | fail→fail | 7,657 | 10,166 | +33% | 1 | 1 | 0% | 912 | 2,406 | +164% | 0 | 0 | — |
case-21 | pass→fail | 15,437 | 7,768 | -50% | 1 | 1 | 0% | 2,396 | 2,244 | -6% | 0 | 0 | — |
case-22 | pass→pass | 11,366 | 3,209 | -72% | 1 | 1 | 0% | 1,731 | 1,405 | -19% | 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 20 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 +36 percentage points is the difference between those two pass rates over the 20 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.