Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Standardized tool set definitions for Claude Code agents ensuring consistent tool access across similar agent types
.claude/skills/aiskillstore-tool-presets/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -33% | 0% |
Standardized tool set definitions for Claude Code agents. Use these presets to ensure consistent tool access across similar agent types.
| Preset | Tools | Best For | |--------|-------|----------| | dev-tools | Read, Write, Edit, Bash | Development/coding agents | | file-ops | Read, Write, Edit, Grep, Glob | File manipulation agents | | analysis | Read, Grep, Glob, Bash | Code analysis agents | | research | Read, Write, WebSearch, WebFetch | Research agents | | orchestration | Read, Write, Edit, Task, TodoWrite | Coordinator agents | | full-stack | All tools | Comprehensive agents |
Reference a preset in your agent's frontmatter:
yaml--- name: my-agent description: Agent description tools: Read, Write, Edit, Bash # Use dev-tools preset pattern skills: tool-presets ---
dev-toolsfile-opsanalysisresearchorchestrationfull-stackFor detailed tool lists per preset, see:
dev-tools.md - Development tools presetfile-ops.md - File operations presetanalysis.md - Code analysis presetresearch.md - Research tools presetorchestration.md - Multi-agent orchestration preset| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 6,745 | 6,041 | -10% | 1 | 1 | 0% | 1,155 | 1,479 | +28% | 0 | 0 | — |
case-02 | fail→pass | 7,332 | 3,321 | -55% | 1 | 1 | 0% | 1,179 | 1,052 | -11% | 0 | 0 | — |
case-03 | fail→pass | 26,667 | 2,852 | -89% | 1 | 1 | 0% | 2,670 | 1,004 | -62% | 0 | 0 | — |
case-04 | fail→pass | 23,602 | 4,568 | -81% | 1 | 1 | 0% | 2,001 | 1,199 | -40% | 0 | 0 | — |
case-14 | fail→pass | 6,875 | 1,891 | -72% | 1 | 1 | 0% | 1,013 | 762 | -25% | 0 | 0 | — |
case-05 | fail→pass | 8,873 | 3,522 | -60% | 1 | 1 | 0% | 1,538 | 1,030 | -33% | 0 | 0 | — |
case-06 | fail→fail | 12,040 | 5,639 | -53% | 1 | 1 | 0% | 1,998 | 1,065 | -47% | 0 | 0 | — |
case-07 | pass→pass | 9,496 | 4,133 | -56% | 1 | 1 | 0% | 1,532 | 1,291 | -16% | 0 | 0 | — |
case-08 | pass→pass | 8,393 | 4,216 | -50% | 1 | 1 | 0% | 1,279 | 1,255 | -2% | 0 | 0 | — |
case-09 | fail→pass | 20,150 | 3,366 | -83% | 1 | 1 | 0% | 2,268 | 995 | -56% | 0 | 0 | — |
case-10 | pass→pass | 8,260 | 2,897 | -65% | 1 | 1 | 0% | 1,295 | 945 | -27% | 0 | 0 | — |
case-11 | pass→pass | 13,103 | 2,956 | -77% | 1 | 1 | 0% | 2,066 | 988 | -52% | 0 | 0 | — |
case-12 | fail→pass | 6,554 | 2,591 | -60% | 1 | 1 | 0% | 1,170 | 829 | -29% | 0 | 0 | — |
case-13 | fail→pass | 5,438 | 2,202 | -60% | 1 | 1 | 0% | 895 | 754 | -16% | 0 | 0 | — |
case-15 | fail→pass | 8,546 | 2,121 | -75% | 1 | 1 | 0% | 1,349 | 822 | -39% | 0 | 0 | — |
case-16 | fail→pass | 9,537 | 2,197 | -77% | 1 | 1 | 0% | 1,540 | 713 | -54% | 0 | 0 | — |
case-17 | fail→pass | 8,278 | 1,758 | -79% | 1 | 1 | 0% | 1,265 | 717 | -43% | 0 | 0 | — |
case-18 | fail→pass | 10,011 | 2,160 | -78% | 1 | 1 | 0% | 1,519 | 734 | -52% | 0 | 0 | — |
case-24 | pass→pass | 13,956 | 17,609 | +26% | 1 | 1 | 0% | 2,748 | 4,337 | +58% | 0 | 0 | — |
case-19 | pass→pass | 8,944 | 3,457 | -61% | 1 | 1 | 0% | 1,409 | 1,059 | -25% | 0 | 0 | — |
case-20 | fail→pass | 11,382 | 3,744 | -67% | 1 | 1 | 0% | 1,826 | 1,101 | -40% | 0 | 0 | — |
case-21 | fail→pass | 12,291 | 2,933 | -76% | 1 | 1 | 0% | 2,059 | 865 | -58% | 0 | 0 | — |
case-22 | pass→fail | 15,526 | 14,020 | -10% | 1 | 1 | 0% | 2,615 | 2,727 | +4% | 0 | 0 | — |
case-23 | pass→pass | 5,009 | 4,226 | -16% | 1 | 1 | 0% | 874 | 1,239 | +42% | 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. 24 cases were attempted. The headline lift of +54 percentage points is the difference between those two pass rates over the 24 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.