Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use the Figma MCP server to fetch design context, screenshots, variables, and assets from Figma, and to translate Figma nodes into production code. Use when a task involves Figma URLs, node IDs, design-to-code implementation, or Figma MCP setup and troubleshooting. Covers general Figma data fetching and exploration. Do NOT use when the goal is specifically pixel-perfect code implementation from a Figma design (use figma-implement-design instead).
.claude/skills/tech-leads-club-figma/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -6% | 0% |
Use the Figma MCP server for Figma-driven implementation. For setup and debugging details (env vars, config, verification), see references/figma-mcp-config.md.
These rules define how to translate Figma inputs into code for this project and must be followed for every Figma-driven change.
references/figma-mcp-config.md — setup, verification, troubleshooting, and link-based usage reminders.references/figma-tools-and-prompts.md — tool catalog and prompt patterns for selecting frameworks/components and fetching metadata.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 9,187 | 58,236 | +534% | 1 | 1 | 0% | 1,965 | 991 | -50% | 0 | 0 | — |
case-02 | fail→fail | 15,228 | 5,779 | -62% | 1 | 1 | 0% | 3,300 | 1,041 | -68% | 0 | 0 | — |
case-03 | fail→fail | 27,532 | 5,963 | -78% | 1 | 1 | 0% | 6,205 | 928 | -85% | 0 | 0 | — |
case-04 | fail→pass | 9,778 | 3,145 | -68% | 1 | 1 | 0% | 1,778 | 1,185 | -33% | 0 | 0 | — |
case-05 | fail→pass | 10,974 | 3,507 | -68% | 1 | 1 | 0% | 1,850 | 1,174 | -37% | 0 | 0 | — |
case-06 | fail→pass | 8,115 | 2,489 | -69% | 1 | 1 | 0% | 1,448 | 1,059 | -27% | 0 | 0 | — |
case-07 | pass→pass | 10,512 | 5,190 | -51% | 1 | 1 | 0% | 1,801 | 1,602 | -11% | 0 | 0 | — |
case-08 | pass→pass | 11,389 | 5,581 | -51% | 1 | 1 | 0% | 1,984 | 1,623 | -18% | 0 | 0 | — |
case-09 | pass→pass | 3,345 | 2,600 | -22% | 1 | 1 | 0% | 514 | 1,096 | +113% | 0 | 0 | — |
case-10 | pass→pass | 8,781 | 3,608 | -59% | 1 | 1 | 0% | 1,430 | 1,157 | -19% | 0 | 0 | — |
case-11 | pass→pass | 8,453 | 5,865 | -31% | 1 | 1 | 0% | 1,870 | 1,733 | -7% | 0 | 0 | — |
case-12 | pass→fail | 10,160 | 3,997 | -61% | 1 | 1 | 0% | 1,813 | 1,346 | -26% | 0 | 0 | — |
case-13 | pass→pass | 9,748 | 4,112 | -58% | 1 | 1 | 0% | 1,696 | 1,480 | -13% | 0 | 0 | — |
case-14 | pass→pass | 12,798 | 2,533 | -80% | 1 | 1 | 0% | 2,427 | 1,066 | -56% | 0 | 0 | — |
case-15 | fail→pass | 10,016 | 4,708 | -53% | 1 | 1 | 0% | 1,748 | 1,506 | -14% | 0 | 0 | — |
case-16 | pass→pass | 14,188 | 11,571 | -18% | 1 | 1 | 0% | 2,552 | 2,633 | +3% | 0 | 0 | — |
case-17 | fail→pass | 9,395 | 3,873 | -59% | 1 | 1 | 0% | 1,488 | 1,399 | -6% | 0 | 0 | — |
case-18 | pass→pass | 8,472 | 2,864 | -66% | 1 | 1 | 0% | 1,505 | 1,140 | -24% | 0 | 0 | — |
case-19 | pass→pass | 10,639 | 2,147 | -80% | 1 | 1 | 0% | 1,720 | 984 | -43% | 0 | 0 | — |
case-20 | pass→pass | 5,422 | 5,533 | +2% | 1 | 1 | 0% | 1,037 | 1,721 | +66% | 0 | 0 | — |
case-21 | pass→pass | 9,103 | 5,836 | -36% | 1 | 1 | 0% | 1,601 | 1,647 | +3% | 0 | 0 | — |
case-22 | pass→pass | 4,490 | 4,341 | -3% | 1 | 1 | 0% | 872 | 1,519 | +74% | 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 19 counted toward the lift figure. The other 3 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 +18 percentage points is the difference between those two pass rates over the 19 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.