Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generates phased, dependency-ordered implementation tasks from specifications. Use after spec is complete and before starting implementation.
.claude/skills/athola-task-planning/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -2% | 0% |
Transforms specifications and implementation plans into concrete, dependency-ordered tasks. Creates phased breakdowns that guide systematic implementation.
Tasks follow a 5-phase structure from setup through polish:
For detailed phase definitions, selection guidelines, and anti-patterns, see modules/phase-structure.md.
Each task includes:
Dependencies define execution order and identify parallelization opportunities:
Nonconflicting Criteria for Parallel Execution:
Mark tasks with P] ONLY if they pass ALL criteria above.
For fan-out/fan-in patterns, task ID conventions, and validation rules, see modules/dependency-patterns.md.
markdown## Phase 2: Core Implementation ### TASK-007 - Implement user authentication service [P] **Dependencies**: TASK-003, TASK-004 **Files**: src/services/auth.ts, src/types/user.ts **Criteria**: All auth tests pass, tokens are valid JWT
Verification: Run pytest -v to verify tests pass.
spec-writing: Creating source specificationsspeckit-orchestrator: Workflow coordinationTASK-NNN format and includesDescription, Phase (0-4), Dependencies list, Parallel marker ([P] or sequential), Files, and Criteria fields
[P] pass all 5 nonconflicting criteria: no fileoverlap, no shared state, dependencies satisfied, no merge conflicts possible, no output dependency on each other
at least one task; unmapped requirements are flagged
one with a higher task number within the same phase
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 29,282 | 40,232 | +37% | 1 | 1 | 0% | 5,559 | 6,262 | +13% | 0 | 0 | — |
case-03 | fail→pass | 23,231 | 23,531 | +1% | 1 | 1 | 0% | 4,122 | 5,278 | +28% | 0 | 0 | — |
case-01 | fail→pass | 23,919 | 23,047 | -4% | 1 | 1 | 0% | 4,205 | 5,365 | +28% | 0 | 0 | — |
case-04 | pass→fail | 33,280 | 32,995 | -1% | 1 | 1 | 0% | 6,179 | 6,971 | +13% | 0 | 0 | — |
case-05 | pass→fail | 18,816 | 41,476 | +120% | 1 | 1 | 0% | 2,995 | 6,964 | +133% | 0 | 0 | — |
case-06 | pass→fail | 17,800 | 20,614 | +16% | 1 | 1 | 0% | 2,986 | 4,475 | +50% | 0 | 0 | — |
case-07 | pass→pass | 7,425 | 5,448 | -27% | 1 | 1 | 0% | 1,414 | 1,750 | +24% | 0 | 0 | — |
case-08 | pass→pass | 6,983 | 4,202 | -40% | 1 | 1 | 0% | 1,130 | 1,482 | +31% | 0 | 0 | — |
case-09 | pass→pass | 3,726 | 5,808 | +56% | 1 | 1 | 0% | 647 | 1,933 | +199% | 0 | 0 | — |
case-10 | fail→pass | 14,489 | 3,310 | -77% | 1 | 1 | 0% | 1,261 | 1,354 | +7% | 0 | 0 | — |
case-11 | fail→pass | 10,000 | 4,960 | -50% | 1 | 1 | 0% | 1,603 | 1,570 | -2% | 0 | 0 | — |
case-12 | pass→pass | 7,994 | 2,817 | -65% | 1 | 1 | 0% | 1,228 | 1,281 | +4% | 0 | 0 | — |
case-13 | fail→pass | 7,002 | 5,618 | -20% | 1 | 1 | 0% | 1,130 | 1,758 | +56% | 0 | 0 | — |
case-14 | pass→pass | 11,766 | 13,305 | +13% | 1 | 1 | 0% | 2,009 | 3,239 | +61% | 0 | 0 | — |
case-15 | fail→pass | 6,905 | 3,106 | -55% | 1 | 1 | 0% | 993 | 1,345 | +35% | 0 | 0 | — |
case-16 | fail→pass | 5,455 | 4,101 | -25% | 1 | 1 | 0% | 890 | 1,461 | +64% | 0 | 0 | — |
case-17 | fail→pass | 6,721 | 4,452 | -34% | 1 | 1 | 0% | 1,227 | 1,602 | +31% | 0 | 0 | — |
case-18 | pass→pass | 4,767 | 4,634 | -3% | 1 | 1 | 0% | 844 | 1,616 | +91% | 0 | 0 | — |
case-19 | fail→pass | 10,317 | 9,326 | -10% | 1 | 1 | 0% | 1,999 | 2,677 | +34% | 0 | 0 | — |
case-20 | fail→pass | 4,573 | 4,015 | -12% | 1 | 1 | 0% | 763 | 1,535 | +101% | 0 | 0 | — |
case-21 | pass→pass | 6,677 | 3,875 | -42% | 1 | 1 | 0% | 1,112 | 1,416 | +27% | 0 | 0 | — |
case-22 | fail→pass | 9,929 | 6,572 | -34% | 1 | 1 | 0% | 1,702 | 1,876 | +10% | 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 +41 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 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.