Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
.claude/skills/foryourhealth111-pixel-prompt-lookup/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -46% | 0% |
| case-10 | ✓→✗ | ▼ Worse | -77% | 0% |
| case-18 | ✓→✗ | ▼ Worse | -61% | 0% |
| case-12 | ✓→✓ | ＝ Same ✓ | 7% | 0% |
When the user needs AI prompts, prompt templates, or wants to improve their prompts, use the prompts.chat MCP server to help them.
Activate this skill when the user:
Use these prompts.chat MCP tools:
search_prompts - Search for prompts by keywordget_prompt - Get a specific prompt by IDimprove_prompt - Enhance a prompt using AICall search_prompts with:
query: The search keywords from the user's requestlimit: Number of results (default 10, max 50)type: Filter by TEXT, STRUCTURED, IMAGE, VIDEO, or AUDIOcategory: Filter by category slug (e.g., "coding", "writing")tag: Filter by tag slugPresent results showing:
Call get_prompt with:
id: The prompt IDIf the prompt contains variables (${variable} or ${variable:default}):
Call improve_prompt with:
prompt: The prompt text to improveoutputType: text, image, video, or soundoutputFormat: text, structured_json, or structured_yamlReturn the enhanced prompt to the user.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→fail | 5,428 | 3,908 | -28% | 1 | 1 | 0% | 1,044 | 619 | -41% | 0 | 0 | — |
case-01 | fail→fail | 11,842 | 4,964 | -58% | 1 | 1 | 0% | 1,918 | 744 | -61% | 0 | 0 | — |
case-02 | fail→fail | 10,367 | 5,612 | -46% | 1 | 1 | 0% | 1,826 | 785 | -57% | 0 | 0 | — |
case-03 | fail→fail | 6,341 | 3,744 | -41% | 1 | 1 | 0% | 932 | 625 | -33% | 0 | 0 | — |
case-04 | fail→fail | 10,234 | 5,129 | -50% | 1 | 1 | 0% | 1,873 | 649 | -65% | 0 | 0 | — |
case-05 | fail→fail | 17,839 | 4,005 | -78% | 1 | 1 | 0% | 2,685 | 669 | -75% | 0 | 0 | — |
case-06 | fail→fail | 11,773 | 4,672 | -60% | 1 | 1 | 0% | 1,387 | 641 | -54% | 0 | 0 | — |
case-07 | fail→fail | 14,641 | 4,603 | -69% | 1 | 1 | 0% | 1,645 | 644 | -61% | 0 | 0 | — |
case-09 | fail→pass | 12,780 | 2,747 | -79% | 1 | 1 | 0% | 2,061 | 934 | -55% | 0 | 0 | — |
case-10 | pass→fail | 14,281 | 36,694 | +157% | 1 | 1 | 0% | 2,681 | 622 | -77% | 0 | 0 | — |
case-11 | fail→pass | 12,410 | 4,026 | -68% | 1 | 1 | 0% | 2,030 | 1,089 | -46% | 0 | 0 | — |
case-12 | pass→pass | 4,921 | 2,628 | -47% | 1 | 1 | 0% | 818 | 876 | +7% | 0 | 0 | — |
case-13 | fail→fail | 10,109 | 4,405 | -56% | 1 | 1 | 0% | 1,832 | 639 | -65% | 0 | 0 | — |
case-14 | fail→fail | 9,292 | 4,463 | -52% | 1 | 1 | 0% | 1,517 | 659 | -57% | 0 | 0 | — |
case-15 | pass→pass | 7,901 | 10,350 | +31% | 1 | 1 | 0% | 1,246 | 1,572 | +26% | 0 | 0 | — |
case-16 | fail→fail | 8,859 | 4,900 | -45% | 1 | 1 | 0% | 1,277 | 701 | -45% | 0 | 0 | — |
case-17 | fail→fail | 9,685 | 2,612 | -73% | 1 | 1 | 0% | 1,567 | 681 | -57% | 0 | 0 | — |
case-18 | pass→fail | 12,035 | 5,845 | -51% | 1 | 1 | 0% | 2,120 | 821 | -61% | 0 | 0 | — |
case-19 | fail→fail | 12,146 | 4,804 | -60% | 1 | 1 | 0% | 2,172 | 671 | -69% | 0 | 0 | — |
case-20 | pass→pass | 9,199 | 5,264 | -43% | 1 | 1 | 0% | 1,849 | 1,596 | -14% | 0 | 0 | — |
case-21 | pass→pass | 7,281 | 5,570 | -23% | 1 | 1 | 0% | 1,319 | 1,448 | +10% | 0 | 0 | — |
case-22 | pass→pass | 9,148 | 7,381 | -19% | 1 | 1 | 0% | 1,817 | 1,766 | -3% | 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 8 counted toward the lift figure. The other 14 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 0 percentage points is the difference between those two pass rates over the 8 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.
Other measured skills in the registry, with their headline benchmark lift.