Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Converts a specification into a phased, dependency-ordered implementation plan. Use after specification is complete and before execution begins.
.claude/skills/athola-project-planning/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -51% | 0% |
For detailed task planning workflows, this skill delegates to spec-kit:task-planning as the canonical implementation. Use this skill for quick planning needs; use spec-kit for thorough project plans.
Skill(attune:project-specification) first)Skill(attune:project-brainstorming) instead)Skill(attune:project-execution) instead)With superpowers:
Skill(superpowers:writing-plans) for structured planningWithout superpowers:
Before completing plan:
Planning is where architecture and scope decisions get made: the pattern chosen, the dependency accepted, the work deferred. Record each decision that had real alternatives to docs/tradeoffs.md (draft and confirm):
Skill(leyline:decision-journal) and appenda tradeoff entry (the decision, the options weighed, and what was sacrificed; set phase to plan). 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/implementation-plan.md is saved, MUST auto-invoke the next phase.
When continuation is invoked:
docs/implementation-plan.md exists and is non-empty Implementation plan complete. Saved to docs/implementation-plan.md. Proceeding to execution phase...
Skill(attune:project-execution)
Bypass Conditions (ONLY skip continuation if ANY true):
--standalone flag was provided by the userdocs/implementation-plan.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/implementation-plan.md exists, is non-empty, and passes QualityChecks (acyclic dependencies, FRs mapped, critical path identified).
docs/tradeoffs.md.
Skill(superpowers:writing-plans) - Planning methodology (if available)Skill(spec-kit:task-planning) - Task breakdown (if available)Skill(attune:project-specification) - Previous phaseSkill(attune:project-execution) - AUTO-INVOKED next phase after planningSkill(attune:mission-orchestrator) - Full lifecycle orchestration/attune:blueprint - Invoke this skill/attune:execute - Next step in workflowSee /attune:blueprint command documentation for complete examples.
If you find circular dependencies in your task graph, break one of the tasks into smaller sub-tasks. If sprint capacity is consistently exceeded, re-estimate tasks using the Fibonacci scale or reduce sprint scope.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 20,393 | 4,907 | -76% | 1 | 1 | 0% | 3,493 | 1,367 | -61% | 0 | 0 | — |
case-02 | fail→fail | 11,008 | 5,255 | -52% | 1 | 1 | 0% | 1,747 | 1,299 | -26% | 0 | 0 | — |
case-03 | fail→fail | 11,841 | 5,545 | -53% | 1 | 1 | 0% | 1,856 | 1,279 | -31% | 0 | 0 | — |
case-04 | fail→pass | 17,424 | 3,722 | -79% | 1 | 1 | 0% | 2,657 | 1,514 | -43% | 0 | 0 | — |
case-05 | fail→pass | 16,194 | 4,076 | -75% | 1 | 1 | 0% | 2,477 | 1,647 | -34% | 0 | 0 | — |
case-06 | fail→fail | 18,158 | 8,176 | -55% | 1 | 1 | 0% | 2,851 | 2,255 | -21% | 0 | 0 | — |
case-16 | fail→pass | 14,738 | 7,167 | -51% | 1 | 1 | 0% | 2,316 | 2,148 | -7% | 0 | 0 | — |
case-07 | fail→fail | 5,571 | 6,070 | +9% | 1 | 1 | 0% | 859 | 1,347 | +57% | 0 | 0 | — |
case-08 | pass→pass | 12,442 | 6,075 | -51% | 1 | 1 | 0% | 1,905 | 1,981 | +4% | 0 | 0 | — |
case-09 | pass→fail | 13,639 | 5,188 | -62% | 1 | 1 | 0% | 2,022 | 1,694 | -16% | 0 | 0 | — |
case-10 | fail→fail | 12,680 | 5,774 | -54% | 1 | 1 | 0% | 1,937 | 1,853 | -4% | 0 | 0 | — |
case-11 | fail→pass | 11,498 | 3,460 | -70% | 1 | 1 | 0% | 1,800 | 1,505 | -16% | 0 | 0 | — |
case-12 | fail→pass | 20,207 | 3,631 | -82% | 1 | 1 | 0% | 3,249 | 1,586 | -51% | 0 | 0 | — |
case-13 | pass→pass | 13,466 | 20,738 | +54% | 1 | 1 | 0% | 2,262 | 4,262 | +88% | 0 | 0 | — |
case-14 | pass→pass | 10,640 | 5,215 | -51% | 1 | 1 | 0% | 1,717 | 1,862 | +8% | 0 | 0 | — |
case-15 | pass→pass | 8,322 | 2,867 | -66% | 1 | 1 | 0% | 1,133 | 1,373 | +21% | 0 | 0 | — |
case-17 | pass→pass | 12,467 | 3,811 | -69% | 1 | 1 | 0% | 2,023 | 1,607 | -21% | 0 | 0 | — |
case-18 | fail→fail | 18,241 | 29,532 | +62% | 1 | 1 | 0% | 2,999 | 6,452 | +115% | 0 | 0 | — |
case-19 | pass→pass | 11,004 | 2,976 | -73% | 1 | 1 | 0% | 1,721 | 1,452 | -16% | 0 | 0 | — |
case-20 | pass→pass | 12,746 | 2,448 | -81% | 1 | 1 | 0% | 2,003 | 1,270 | -37% | 0 | 0 | — |
case-21 | fail→fail | 10,161 | 2,850 | -72% | 1 | 1 | 0% | 1,425 | 1,444 | +1% | 0 | 0 | — |
case-22 | fail→pass | 4,699 | 1,047 | -78% | 1 | 1 | 0% | 634 | 1,092 | +72% | 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 +23 percentage points is the difference between those two pass rates over the 18 comparable cases. 1 case got worse with the skill loaded, and it is 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.