Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Import a GSD-2 (.gsd/) project back to GSD v1 (.planning/) format
.claude/skills/davepoon-gsd-from-gsd2/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-22 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -30% | 0% |
<objective> Reverse-migrate a GSD-2 project (.gsd/ directory) back to GSD v1 (.planning/) format.
Maps the GSD-2 hierarchy (Milestone → Slice → Task) to the GSD v1 hierarchy (Milestone sections in ROADMAP.md → Phase → Plan), preserving completion state, research files, and summaries.
CJS-only: from-gsd2 is not on the gsd-sdk query registry; call gsd-tools.cjs as shown below (see docs/CLI-TOOLS.md). </objective>
<process>
--path argument):bash node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" from-gsd2 --dry-run If no .gsd/ is found, report the error and stop.
bash node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" from-gsd2 Use --force if .planning/ already exists and the user has confirmed overwrite.
filesWritten count, planningDir path, and the preview summary.</process>
<notes>
.gsd/ is never modified or removed.--path <dir> to migrate a project at a different path than the current directory.</notes>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 2,498 | 2,552 | +2% | 1 | 1 | 0% | 276 | 837 | +203% | 0 | 0 | — |
case-02 | fail→fail | 10,531 | 2,508 | -76% | 1 | 1 | 0% | 2,098 | 857 | -59% | 0 | 0 | — |
case-03 | fail→fail | 9,385 | 2,505 | -73% | 1 | 1 | 0% | 1,691 | 844 | -50% | 0 | 0 | — |
case-08 | fail→pass | 8,550 | 1,847 | -78% | 1 | 1 | 0% | 1,546 | 859 | -44% | 0 | 0 | — |
case-04 | fail→pass | 6,511 | 3,438 | -47% | 1 | 1 | 0% | 1,184 | 1,166 | -2% | 0 | 0 | — |
case-05 | pass→pass | 9,841 | 4,172 | -58% | 1 | 1 | 0% | 1,910 | 1,402 | -27% | 0 | 0 | — |
case-06 | pass→pass | 14,228 | 4,647 | -67% | 1 | 1 | 0% | 2,896 | 1,538 | -47% | 0 | 0 | — |
case-07 | fail→pass | 9,040 | 5,733 | -37% | 1 | 1 | 0% | 1,603 | 1,536 | -4% | 0 | 0 | — |
case-22 | fail→pass | 11,101 | 4,570 | -59% | 1 | 1 | 0% | 2,007 | 1,357 | -32% | 0 | 0 | — |
case-09 | fail→pass | 8,535 | 3,212 | -62% | 1 | 1 | 0% | 1,590 | 1,106 | -30% | 0 | 0 | — |
case-10 | fail→pass | 8,676 | 4,682 | -46% | 1 | 1 | 0% | 1,607 | 1,419 | -12% | 0 | 0 | — |
case-11 | fail→pass | 9,565 | 4,544 | -52% | 1 | 1 | 0% | 1,610 | 1,258 | -22% | 0 | 0 | — |
case-12 | fail→pass | 8,481 | 2,739 | -68% | 1 | 1 | 0% | 1,547 | 1,023 | -34% | 0 | 0 | — |
case-13 | pass→pass | 13,225 | 5,276 | -60% | 1 | 1 | 0% | 2,404 | 1,579 | -34% | 0 | 0 | — |
case-14 | fail→pass | 10,584 | 2,437 | -77% | 1 | 1 | 0% | 2,061 | 946 | -54% | 0 | 0 | — |
case-15 | fail→pass | 11,115 | 2,833 | -75% | 1 | 1 | 0% | 1,981 | 1,045 | -47% | 0 | 0 | — |
case-16 | pass→pass | 10,810 | 3,960 | -63% | 1 | 1 | 0% | 1,638 | 1,218 | -26% | 0 | 0 | — |
case-17 | pass→pass | 8,087 | 2,426 | -70% | 1 | 1 | 0% | 1,308 | 944 | -28% | 0 | 0 | — |
case-18 | pass→pass | 4,215 | 2,727 | -35% | 1 | 1 | 0% | 875 | 1,096 | +25% | 0 | 0 | — |
case-19 | pass→pass | 5,159 | 4,083 | -21% | 1 | 1 | 0% | 1,030 | 1,417 | +38% | 0 | 0 | — |
case-20 | fail→pass | 9,484 | 5,225 | -45% | 1 | 1 | 0% | 1,713 | 1,467 | -14% | 0 | 0 | — |
case-21 | fail→pass | 11,714 | 5,196 | -56% | 1 | 1 | 0% | 2,082 | 1,507 | -28% | 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 +55 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.