Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when a problem or issue needs explicit framing before deeper analysis begins — establishing the Problem statement, Current state, Desired state, Stakeholders, and Success criteria. This should trigger when an issue's Problem Framing point of view needs evaluation, or when a maintainer directly asks to frame a problem before root-cause analysis, design, or planning begins. Part of Plinth Toolkit
.claude/skills/jabrena-021-problem-framing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 26% | 0% |
Guide production of a Problem statement, Current state, Desired state, Stakeholders, and Success criteria for a problem or issue under exploration. This is an interactive SKILL.
What is covered in this Skill?
022-root-cause-analysis and the remaining Functional Specification lensesFrame the problem before any deeper analysis. When this technique is orchestrated by another workflow, the orchestrator owns clarifying-question sequencing; when applied standalone, ask directly.
references/021-problem-framing.md before applying Problem Framing guidanceRead references/021-problem-framing.md, then identify what evidence is already available for the problem statement, current state, desired state, stakeholders, and success criteria.
Strip out wording that already assumes a fix, technology, or implementation, and restate the gap in observable terms.
Describe the current state as observable facts and the desired state as a target outcome or capability.
List stakeholders affected by, accountable for, or informed about the problem, then define observable or measurable success criteria.
Report the Problem statement, Current state, Desired state, Stakeholders, and Success criteria, and flag any field left open pending a clarifying answer.
For detailed guidance, examples, and constraints, see references/021-problem-framing.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,041 | 12,547 | +4% | 1 | 1 | 0% | 1,871 | 2,640 | +41% | 0 | 0 | — |
case-02 | fail→pass | 13,350 | 11,011 | -18% | 1 | 1 | 0% | 2,125 | 2,354 | +11% | 0 | 0 | — |
case-03 | fail→fail | 12,169 | 10,611 | -13% | 1 | 1 | 0% | 1,815 | 2,250 | +24% | 0 | 0 | — |
case-04 | pass→pass | 2,612 | 6,201 | +137% | 1 | 1 | 0% | 469 | 1,656 | +253% | 0 | 0 | — |
case-05 | pass→pass | 2,748 | 3,584 | +30% | 1 | 1 | 0% | 481 | 1,256 | +161% | 0 | 0 | — |
case-06 | pass→pass | 3,058 | 6,698 | +119% | 1 | 1 | 0% | 554 | 1,834 | +231% | 0 | 0 | — |
case-07 | fail→pass | 13,149 | 19,697 | +50% | 1 | 1 | 0% | 2,106 | 2,131 | +1% | 0 | 0 | — |
case-08 | fail→pass | 14,697 | 32,852 | +124% | 1 | 1 | 0% | 2,322 | 2,412 | +4% | 0 | 0 | — |
case-09 | fail→fail | 22,847 | 12,241 | -46% | 1 | 1 | 0% | 2,535 | 2,283 | -10% | 0 | 0 | — |
case-10 | fail→pass | 12,595 | 8,660 | -31% | 1 | 1 | 0% | 1,596 | 2,010 | +26% | 0 | 0 | — |
case-11 | fail→pass | 14,835 | 11,795 | -20% | 1 | 1 | 0% | 2,318 | 2,425 | +5% | 0 | 0 | — |
case-12 | pass→pass | 15,474 | 11,146 | -28% | 1 | 1 | 0% | 2,400 | 2,405 | +0% | 0 | 0 | — |
case-13 | fail→pass | 11,536 | 9,958 | -14% | 1 | 1 | 0% | 1,773 | 2,195 | +24% | 0 | 0 | — |
case-14 | pass→pass | 14,833 | 8,640 | -42% | 1 | 1 | 0% | 2,396 | 2,003 | -16% | 0 | 0 | — |
case-15 | pass→pass | 16,069 | 10,597 | -34% | 1 | 1 | 0% | 2,640 | 2,300 | -13% | 0 | 0 | — |
case-16 | fail→pass | 11,937 | 8,342 | -30% | 1 | 1 | 0% | 1,773 | 1,829 | +3% | 0 | 0 | — |
case-17 | fail→pass | 12,568 | 9,379 | -25% | 1 | 1 | 0% | 2,074 | 2,145 | +3% | 0 | 0 | — |
case-18 | fail→pass | 16,028 | 14,844 | -7% | 1 | 1 | 0% | 2,390 | 2,993 | +25% | 0 | 0 | — |
case-19 | fail→pass | 13,108 | 8,806 | -33% | 1 | 1 | 0% | 2,041 | 2,019 | -1% | 0 | 0 | — |
case-20 | fail→pass | 22,004 | 8,633 | -61% | 1 | 1 | 0% | 3,334 | 1,890 | -43% | 0 | 0 | — |
case-21 | fail→pass | 12,952 | 9,221 | -29% | 1 | 1 | 0% | 1,945 | 2,003 | +3% | 0 | 0 | — |
case-22 | fail→pass | 14,243 | 11,081 | -22% | 1 | 1 | 0% | 2,141 | 2,235 | +4% | 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 +64 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.