Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Query Pollinations text API with web-search models (gemini-search, perplexity-fast, nomnom, etc.). Use when you need web search grounded answers via Pollinations.
.claude/skills/pollinations-web-research/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -38% | 0% |
By default, queries both gemini-search and perplexity-fast in parallel.
curlSet an API key in an environment variable (preferred):
bashexport POLLINATIONS_API_KEY="YOUR_KEY"
Or create a local .env file at .claude/skills/web-research/.env:
bashPOLLINATIONS_API_KEY="YOUR_KEY"
If POLLINATIONS_API_KEY is not set, the script will prompt for a key (input hidden).
bash.claude/skills/web-research/scripts/web-research.sh "What is pollinations.ai?"
Choose a model:
bash.claude/skills/web-research/scripts/web-research.sh --model perplexity-fast "Fact-check this claim with sources"
Compare multiple models:
bash.claude/skills/web-research/scripts/web-research.sh --models gemini-search,perplexity-fast,nomnom "Compare answers"
Run multi-model in parallel:
bash.claude/skills/web-research/scripts/web-research.sh --models gemini-search,perplexity-fast --parallel "Compare answers"
Run two searches first — "<vendor> <capability> <year>" and "<capability> MCP server" — and record the result. A first-party vendor launch in the last two quarters gets its own section ("why not the managed product"). It's the cheapest part of the review and can invalidate the whole plan, so do it before studying the design.
A summarizer given a leading prompt ("did X beat Y?") tends to echo the prompt, including invented quotes. Ask neutrally ("list every numeric result table") and, for any number you'll repeat, read the source text or table yourself.
https://gen.pollinations.ai/v1/chat/completionsAuthorization: Bearer <key>| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 41,677 | 23,178 | -44% | 1 | 1 | 0% | 3,063 | 4,058 | +32% | 0 | 0 | — |
case-02 | fail→fail | 48,866 | 46,430 | -5% | 1 | 1 | 0% | 549 | 967 | +76% | 0 | 0 | — |
case-03 | fail→fail | 18,714 | 30,706 | +64% | 1 | 1 | 0% | 556 | 2,032 | +265% | 0 | 0 | — |
case-04 | fail→fail | 27,767 | 20,029 | -28% | 1 | 1 | 0% | 484 | 842 | +74% | 0 | 0 | — |
case-05 | fail→fail | 59,359 | 14,327 | -76% | 1 | 1 | 0% | 2,093 | 858 | -59% | 0 | 0 | — |
case-06 | fail→fail | 28,603 | 10,419 | -64% | 1 | 1 | 0% | 3,865 | 824 | -79% | 0 | 0 | — |
case-07 | fail→fail | 12,167 | 11,300 | -7% | 1 | 1 | 0% | 1,511 | 780 | -48% | 0 | 0 | — |
case-08 | fail→fail | 4,428 | 17,254 | +290% | 1 | 1 | 0% | 502 | 762 | +52% | 0 | 0 | — |
case-09 | fail→pass | 15,798 | 11,628 | -26% | 1 | 1 | 0% | 2,153 | 2,128 | -1% | 0 | 0 | — |
case-10 | fail→pass | 16,733 | 12,984 | -22% | 1 | 1 | 0% | 2,114 | 2,195 | +4% | 0 | 0 | — |
case-11 | fail→pass | 49,007 | 6,560 | -87% | 1 | 1 | 0% | 2,484 | 1,110 | -55% | 0 | 0 | — |
case-12 | fail→pass | 25,231 | 5,736 | -77% | 1 | 1 | 0% | 1,757 | 1,090 | -38% | 0 | 0 | — |
case-13 | fail→pass | 10,311 | 2,637 | -74% | 1 | 1 | 0% | 1,295 | 842 | -35% | 0 | 0 | — |
case-14 | fail→pass | 8,795 | 2,648 | -70% | 1 | 1 | 0% | 994 | 742 | -25% | 0 | 0 | — |
case-15 | pass→pass | 11,221 | 2,950 | -74% | 1 | 1 | 0% | 1,462 | 863 | -41% | 0 | 0 | — |
case-16 | pass→pass | 6,131 | 2,397 | -61% | 1 | 1 | 0% | 548 | 739 | +35% | 0 | 0 | — |
case-17 | fail→pass | 44,489 | 4,151 | -91% | 1 | 1 | 0% | 1,461 | 964 | -34% | 0 | 0 | — |
case-18 | fail→pass | 7,278 | 7,262 | -0% | 1 | 1 | 0% | 1,058 | 1,413 | +34% | 0 | 0 | — |
case-19 | fail→pass | 10,234 | 5,570 | -46% | 1 | 1 | 0% | 1,388 | 1,110 | -20% | 0 | 0 | — |
case-20 | fail→pass | 10,401 | 3,070 | -70% | 1 | 1 | 0% | 1,401 | 896 | -36% | 0 | 0 | — |
case-21 | fail→pass | 36,311 | 3,194 | -91% | 1 | 1 | 0% | 1,602 | 796 | -50% | 0 | 0 | — |
case-22 | fail→pass | 17,696 | 3,581 | -80% | 1 | 1 | 0% | 2,616 | 838 | -68% | 0 | 0 | — |
case-23 | pass→pass | 7,680 | 9,291 | +21% | 1 | 1 | 0% | 1,255 | 2,189 | +74% | 0 | 0 | — |
case-24 | pass→pass | 14,644 | 7,492 | -49% | 1 | 1 | 0% | 1,451 | 1,465 | +1% | 0 | 0 | — |
case-25 | pass→pass | 7,230 | 8,055 | +11% | 1 | 1 | 0% | 1,296 | 1,918 | +48% | 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 17 counted toward the lift figure. The other 8 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 +52 percentage points is the difference between those two pass rates over the 17 comparable cases.
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 | 8/10/2026 | +61% |
Other measured skills in the registry, with their headline benchmark lift.