Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you have a written implementation plan to execute in a separate session with review checkpoints
.claude/skills/sickn33-executing-plans/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 81% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -7% | 0% |
Load plan, review critically, execute tasks in batches, report for review between batches.
Core principle: Batch execution with checkpoints for architect review.
Announce at start: "I'm using the executing-plans skill to implement this plan."
Default: First 3 tasks
For each task:
When batch complete:
Based on feedback:
After all tasks complete and verified:
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to Review (Step 1) when:
Don't force through blockers - stop and ask.
This skill is applicable to execute the workflow or actions described in the overview.
User request:
> Use @executing-plans for this task: Use when you have a written implementation plan to execute in a separate session with review checkpoints.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | pass→fail | 9,384 | 3,067 | -67% | 1 | 1 | 0% | 1,687 | 899 | -47% | 0 | 0 | — |
case-01 | fail→fail | 4,646 | 2,328 | -50% | 1 | 1 | 0% | 248 | 884 | +256% | 0 | 0 | — |
case-02 | fail→fail | 4,426 | 2,090 | -53% | 1 | 1 | 0% | 387 | 845 | +118% | 0 | 0 | — |
case-03 | fail→fail | 5,103 | 2,609 | -49% | 1 | 1 | 0% | 323 | 866 | +168% | 0 | 0 | — |
case-04 | fail→pass | 7,587 | 3,488 | -54% | 1 | 1 | 0% | 1,288 | 1,169 | -9% | 0 | 0 | — |
case-05 | fail→pass | 5,191 | 4,823 | -7% | 1 | 1 | 0% | 762 | 1,376 | +81% | 0 | 0 | — |
case-06 | pass→pass | 7,194 | 4,509 | -37% | 1 | 1 | 0% | 1,236 | 1,302 | +5% | 0 | 0 | — |
case-07 | fail→fail | 9,749 | 3,191 | -67% | 1 | 1 | 0% | 1,624 | 942 | -42% | 0 | 0 | — |
case-08 | fail→pass | 4,994 | 2,255 | -55% | 1 | 1 | 0% | 877 | 963 | +10% | 0 | 0 | — |
case-09 | pass→pass | 4,309 | 3,386 | -21% | 1 | 1 | 0% | 746 | 1,202 | +61% | 0 | 0 | — |
case-10 | fail→fail | 5,042 | 2,399 | -52% | 1 | 1 | 0% | 791 | 957 | +21% | 0 | 0 | — |
case-11 | pass→pass | 5,721 | 2,990 | -48% | 1 | 1 | 0% | 873 | 1,050 | +20% | 0 | 0 | — |
case-12 | pass→fail | 2,411 | 1,788 | -26% | 1 | 1 | 0% | 396 | 811 | +105% | 0 | 0 | — |
case-13 | fail→fail | 4,818 | 2,535 | -47% | 1 | 1 | 0% | 722 | 914 | +27% | 0 | 0 | — |
case-14 | pass→pass | 6,391 | 2,180 | -66% | 1 | 1 | 0% | 896 | 882 | -2% | 0 | 0 | — |
case-15 | fail→fail | 8,622 | 2,525 | -71% | 1 | 1 | 0% | 1,207 | 939 | -22% | 0 | 0 | — |
case-16 | fail→pass | 3,476 | 1,266 | -64% | 1 | 1 | 0% | 620 | 766 | +24% | 0 | 0 | — |
case-17 | fail→pass | 9,691 | 4,283 | -56% | 1 | 1 | 0% | 1,399 | 1,297 | -7% | 0 | 0 | — |
case-18 | pass→pass | 11,541 | 3,350 | -71% | 1 | 1 | 0% | 1,343 | 1,058 | -21% | 0 | 0 | — |
case-19 | fail→pass | 8,960 | 3,304 | -63% | 1 | 1 | 0% | 1,243 | 1,059 | -15% | 0 | 0 | — |
case-20 | pass→pass | 49,073 | 34,704 | -29% | 1 | 1 | 0% | 6,180 | 6,157 | -0% | 0 | 0 | — |
case-21 | fail→pass | 12,087 | 14,820 | +23% | 1 | 1 | 0% | 1,232 | 2,224 | +81% | 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 21 counted toward the lift figure. The other 1 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 21 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.