Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Transforms project briefs into testable specifications with user stories and acceptance criteria. Use after brainstorming, before planning.
.claude/skills/athola-project-specification/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-22 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 12% | 0% |
For detailed specification writing workflows, this skill delegates to spec-kit:spec-writing as the canonical implementation. Use this skill for quick specification needs; use spec-kit for thorough specification documents.
Skill(attune:project-brainstorming) instead)Skill(attune:project-planning) instead)Skill(attune:war-room) for complex choices)With spec-kit:
Skill(spec-kit:spec-writing) for methodologyWithout spec-kit:
If the specification settled a design decision with real alternatives (a chosen data model, an interface boundary, a constraint accepted), record it to docs/tradeoffs.md before continuing (draft and confirm):
Skill(leyline:decision-journal) and appenda tradeoff entry (the decision, the options weighed, and what was sacrificed; set phase to specify). Show the draft; append on confirmation.
docs/tradeoffs.md using the in-fileENTRY TEMPLATE; assign the next TR-NNN id.
Automatic Trigger: After Quality Checks pass and docs/specification.md is saved, MUST auto-invoke the next phase.
When continuation is invoked:
docs/specification.md exists and is non-empty Specification complete. Saved to docs/specification.md. Proceeding to planning phase...
Skill(attune:project-planning)
Bypass Conditions (ONLY skip continuation if ANY true):
--standalone flag was provided by the userdocs/specification.md does not exist or is empty (phase failed)Do NOT prompt the user for confirmation: this is a lightweight checkpoint, not an interactive gate. The user can always interrupt if needed.
docs/specification.md exists, is non-empty, and passes Quality Checks.docs/tradeoffs.md (or there was no meaningful design fork).
Skill(spec-kit:spec-writing) - Spec-kit methodology (if available)Skill(attune:project-brainstorming) - Previous phaseSkill(attune:project-planning) - AUTO-INVOKED next phase after specificationSkill(attune:mission-orchestrator) - Full lifecycle orchestration/attune:specify - Invoke this skill/attune:specify --clarify - Run clarification workflow/attune:blueprint - Next step in workflowSee /attune:specify command documentation for complete examples.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | fail→pass | 9,063 | 2,956 | -67% | 1 | 1 | 0% | 1,250 | 1,267 | +1% | 0 | 0 | — |
case-01 | fail→fail | 18,171 | 7,210 | -60% | 1 | 1 | 0% | 3,036 | 1,168 | -62% | 0 | 0 | — |
case-02 | fail→fail | 38,154 | 7,179 | -81% | 1 | 1 | 0% | 5,205 | 1,636 | -69% | 0 | 0 | — |
case-03 | fail→fail | 16,384 | 4,432 | -73% | 1 | 1 | 0% | 2,751 | 1,115 | -59% | 0 | 0 | — |
case-04 | fail→pass | 12,181 | 4,476 | -63% | 1 | 1 | 0% | 1,876 | 1,498 | -20% | 0 | 0 | — |
case-05 | fail→fail | 18,806 | 3,496 | -81% | 1 | 1 | 0% | 3,355 | 1,261 | -62% | 0 | 0 | — |
case-06 | fail→pass | 20,080 | 7,930 | -61% | 1 | 1 | 0% | 2,981 | 1,979 | -34% | 0 | 0 | — |
case-07 | fail→fail | 25,027 | 21,456 | -14% | 1 | 1 | 0% | 4,685 | 4,343 | -7% | 0 | 0 | — |
case-08 | fail→fail | 12,262 | 4,561 | -63% | 1 | 1 | 0% | 2,079 | 1,108 | -47% | 0 | 0 | — |
case-09 | fail→fail | 35,628 | 26,658 | -25% | 1 | 1 | 0% | 6,160 | 5,374 | -13% | 0 | 0 | — |
case-10 | fail→fail | 18,474 | 20,942 | +13% | 1 | 1 | 0% | 3,021 | 3,309 | +10% | 0 | 0 | — |
case-11 | fail→pass | 24,725 | 9,594 | -61% | 1 | 1 | 0% | 4,204 | 2,558 | -39% | 0 | 0 | — |
case-12 | pass→fail | 22,208 | 4,930 | -78% | 1 | 1 | 0% | 3,934 | 1,123 | -71% | 0 | 0 | — |
case-13 | pass→pass | 17,372 | 13,424 | -23% | 1 | 1 | 0% | 2,876 | 3,079 | +7% | 0 | 0 | — |
case-14 | fail→fail | 20,686 | 6,778 | -67% | 1 | 1 | 0% | 3,317 | 1,130 | -66% | 0 | 0 | — |
case-15 | fail→fail | 9,803 | 13,797 | +41% | 1 | 1 | 0% | 1,476 | 2,379 | +61% | 0 | 0 | — |
case-16 | pass→fail | 12,930 | 4,538 | -65% | 1 | 1 | 0% | 2,131 | 1,002 | -53% | 0 | 0 | — |
case-17 | pass→pass | 14,396 | 17,169 | +19% | 1 | 1 | 0% | 2,273 | 3,679 | +62% | 0 | 0 | — |
case-18 | fail→pass | 18,808 | 19,242 | +2% | 1 | 1 | 0% | 3,054 | 3,425 | +12% | 0 | 0 | — |
case-19 | fail→fail | 17,290 | 14,715 | -15% | 1 | 1 | 0% | 2,739 | 2,777 | +1% | 0 | 0 | — |
case-20 | pass→fail | 8,842 | 4,451 | -50% | 1 | 1 | 0% | 1,413 | 1,061 | -25% | 0 | 0 | — |
case-21 | fail→pass | 11,344 | 9,955 | -12% | 1 | 1 | 0% | 1,868 | 2,565 | +37% | 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 15 counted toward the lift figure. The other 7 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 +14 percentage points is the difference between those two pass rates over the 15 comparable cases. 4 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.