Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Update a markdown file section with an index/table of files from a specified folder.
.claude/skills/update-markdown-file-index/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 629% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 889% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 287% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 721% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 37% | 0% |
Update markdown file ${file} with an index/table of files from folder ${input:folder}.
${file} to understand existing structure${input:folder} matching pattern ${input:pattern}For each discovered file, extract:
.md, .js, .py)Choose format based on file types and existing content:
markdown## Files in ${folder} - [filename.ext](path/to/filename.ext) - Description - [filename2.ext](path/to/filename2.ext) - Description
markdown| File | Type | Description | |------|------|-------------| | [filename.ext](path/to/filename.ext) | Extension | Description | | [filename2.ext](path/to/filename2.ext) | Extension | Description |
Group files by type/category with separate sections or sub-tables.
Look for existing sections with these patterns:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 2,304 | 8,454 | +267% | 1 | 1 | 0% | 330 | 2,407 | +629% | 0 | 0 | — |
case-02 | fail→fail | 2,134 | 4,920 | +131% | 1 | 1 | 0% | 282 | 1,042 | +270% | 0 | 0 | — |
case-03 | fail→pass | 4,184 | 7,492 | +79% | 1 | 1 | 0% | 216 | 2,136 | +889% | 0 | 0 | — |
case-04 | fail→pass | 2,578 | 5,805 | +125% | 1 | 1 | 0% | 295 | 1,142 | +287% | 0 | 0 | — |
case-05 | pass→fail | 5,769 | 3,729 | -35% | 1 | 1 | 0% | 1,041 | 729 | -30% | 0 | 0 | — |
case-10 | fail→pass | 4,128 | 7,702 | +87% | 1 | 1 | 0% | 248 | 2,036 | +721% | 0 | 0 | — |
case-06 | pass→pass | 9,706 | 5,945 | -39% | 1 | 1 | 0% | 2,158 | 1,724 | -20% | 0 | 0 | — |
case-07 | pass→pass | 3,418 | 4,823 | +41% | 1 | 1 | 0% | 630 | 1,490 | +137% | 0 | 0 | — |
case-08 | fail→pass | 6,901 | 5,337 | -23% | 1 | 1 | 0% | 1,289 | 1,764 | +37% | 0 | 0 | — |
case-09 | fail→pass | 1,946 | 5,495 | +182% | 1 | 1 | 0% | 339 | 1,773 | +423% | 0 | 0 | — |
case-11 | pass→pass | 3,283 | 4,306 | +31% | 1 | 1 | 0% | 613 | 1,449 | +136% | 0 | 0 | — |
case-12 | pass→pass | 7,761 | 6,070 | -22% | 1 | 1 | 0% | 1,691 | 1,910 | +13% | 0 | 0 | — |
case-13 | fail→pass | 13,579 | 4,352 | -68% | 1 | 1 | 0% | 2,684 | 1,524 | -43% | 0 | 0 | — |
case-14 | pass→pass | 6,815 | 6,358 | -7% | 1 | 1 | 0% | 1,254 | 1,933 | +54% | 0 | 0 | — |
case-15 | pass→pass | 5,443 | 3,661 | -33% | 1 | 1 | 0% | 1,069 | 1,262 | +18% | 0 | 0 | — |
case-16 | pass→pass | 8,673 | 5,443 | -37% | 1 | 1 | 0% | 1,836 | 1,716 | -7% | 0 | 0 | — |
case-17 | fail→pass | 2,390 | 4,711 | +97% | 1 | 1 | 0% | 312 | 1,618 | +419% | 0 | 0 | — |
case-18 | fail→pass | 2,105 | 8,185 | +289% | 1 | 1 | 0% | 296 | 2,134 | +621% | 0 | 0 | — |
case-19 | pass→pass | 2,686 | 4,414 | +64% | 1 | 1 | 0% | 520 | 1,440 | +177% | 0 | 0 | — |
case-20 | pass→pass | 7,346 | 6,696 | -9% | 1 | 1 | 0% | 1,742 | 1,920 | +10% | 0 | 0 | — |
case-21 | pass→fail | 2,466 | 2,511 | +2% | 1 | 1 | 0% | 492 | 1,126 | +129% | 0 | 0 | — |
case-22 | fail→pass | 2,581 | 4,666 | +81% | 1 | 1 | 0% | 393 | 1,551 | +295% | 0 | 0 | — |
case-23 | fail→fail | 2,582 | 1,870 | -28% | 1 | 1 | 0% | 326 | 959 | +194% | 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. 23 cases were attempted, and 21 counted toward the lift figure. The other 2 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 +35 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.
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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/24/2026 | +14% |
Other measured skills in the registry, with their headline benchmark lift.