Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate shell files.
.claude/skills/griddynamics-init-workspace-shells/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -46% | 0% |
<init_workspace_shells>
<role> Shell configuration specialist for IDE/CodingAgent workspace bootstrapping </role>
<when_to_use_skill> Shell files delegate logic to KB via ACQUIRE, enabling centralized instruction updates across projects. Use when onboarding or reconfiguring IDE/Agent workspace shell files. </when_to_use_skill>
<core_concepts>
</core_concepts>
<process>
Internal knowledge about IDE/agent shell configuration is obsolete — LIST and ACQUIRE from KB.
Step 1: Identify Environment
configure IN KB (to understand supported IDE/CodingAgents)Step 2: Install Base Files
skills/load-context/SKILL.md FROM KB — install as SKILLrules/bootstrap.md FROM KB — install as CORE RULE, copy content (no refs/links)Step 3: MUST Generate Skill Shells
skills IN KB with XML formatskill-shell.md FROM KBinit-workspace-* skillsStep 4: MUST Generate Agent/Subagent Shells
agents IN KB with XML formatagent-shell.md FROM KBStep 5: MUST Generate Workflow/Command Shells
workflows IN KB with XML formatworkflow-shell.md FROM KBinit-workspace-* workflows and its phasesStep 6: Verify Shell Integrity
</process>
<validation_checklist>
</validation_checklist>
</init_workspace_shells>
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 5,906 | 7,194 | +22% | 1 | 1 | 0% | 224 | 1,840 | +721% | 0 | 0 | — |
case-02 | fail→fail | 4,391 | 7,993 | +82% | 1 | 1 | 0% | 634 | 1,332 | +110% | 0 | 0 | — |
case-03 | fail→fail | 15,627 | 3,444 | -78% | 1 | 1 | 0% | 2,800 | 924 | -67% | 0 | 0 | — |
case-04 | pass→fail | 24,142 | 30,960 | +28% | 1 | 1 | 0% | 4,421 | 6,774 | +53% | 0 | 0 | — |
case-05 | pass→pass | 24,788 | 19,480 | -21% | 1 | 1 | 0% | 3,603 | 3,653 | +1% | 0 | 0 | — |
case-06 | pass→pass | 20,804 | 14,412 | -31% | 1 | 1 | 0% | 3,331 | 2,959 | -11% | 0 | 0 | — |
case-11 | fail→pass | 9,046 | 3,489 | -61% | 1 | 1 | 0% | 1,466 | 1,064 | -27% | 0 | 0 | — |
case-07 | fail→pass | 6,720 | 4,163 | -38% | 1 | 1 | 0% | 1,058 | 1,250 | +18% | 0 | 0 | — |
case-08 | pass→pass | 15,335 | 8,069 | -47% | 1 | 1 | 0% | 2,459 | 1,870 | -24% | 0 | 0 | — |
case-09 | pass→pass | 14,356 | 2,909 | -80% | 1 | 1 | 0% | 1,977 | 1,067 | -46% | 0 | 0 | — |
case-10 | fail→pass | 11,536 | 3,627 | -69% | 1 | 1 | 0% | 1,688 | 1,118 | -34% | 0 | 0 | — |
case-12 | fail→pass | 10,689 | 2,798 | -74% | 1 | 1 | 0% | 1,677 | 971 | -42% | 0 | 0 | — |
case-13 | pass→pass | 10,530 | 4,496 | -57% | 1 | 1 | 0% | 1,740 | 1,212 | -30% | 0 | 0 | — |
case-14 | fail→pass | 12,001 | 3,067 | -74% | 1 | 1 | 0% | 1,869 | 1,010 | -46% | 0 | 0 | — |
case-15 | fail→pass | 15,527 | 6,139 | -60% | 1 | 1 | 0% | 2,394 | 1,455 | -39% | 0 | 0 | — |
case-16 | fail→pass | 9,240 | 4,972 | -46% | 1 | 1 | 0% | 1,300 | 1,334 | +3% | 0 | 0 | — |
case-17 | pass→pass | 15,242 | 5,208 | -66% | 1 | 1 | 0% | 2,298 | 1,295 | -44% | 0 | 0 | — |
case-18 | fail→pass | 7,340 | 1,816 | -75% | 1 | 1 | 0% | 1,051 | 890 | -15% | 0 | 0 | — |
case-19 | fail→pass | 6,732 | 2,817 | -58% | 1 | 1 | 0% | 1,064 | 948 | -11% | 0 | 0 | — |
case-20 | fail→pass | 9,619 | 4,362 | -55% | 1 | 1 | 0% | 1,367 | 1,194 | -13% | 0 | 0 | — |
case-21 | fail→pass | 9,188 | 2,430 | -74% | 1 | 1 | 0% | 1,317 | 901 | -32% | 0 | 0 | — |
case-22 | fail→pass | 7,301 | 2,014 | -72% | 1 | 1 | 0% | 703 | 921 | +31% | 0 | 0 | — |
case-23 | fail→pass | 9,981 | 1,933 | -81% | 1 | 1 | 0% | 1,476 | 891 | -40% | 0 | 0 | — |
case-24 | fail→pass | 14,460 | 3,633 | -75% | 1 | 1 | 0% | 2,194 | 1,109 | -49% | 0 | 0 | — |
case-25 | fail→pass | 15,857 | 5,893 | -63% | 1 | 1 | 0% | 2,564 | 1,572 | -39% | 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. 25 cases were attempted, and 24 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 +56 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.