Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Calculate NFT rarity scores and rank tokens by trait uniqueness. Use when analyzing NFT collections, checking token rarity, or comparing NFTs. Trigger with phrases like "check NFT rarity", "analyze collection", "rank tokens", "compare NFTs".
.claude/skills/jeremylongshore-analyzing-nft-rarity/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -36% | 0% |
NFT rarity analysis skill that:
OPENSEA_API_KEY for higher rate limitsALCHEMY_API_KEY for direct metadata fetchingbashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py collection boredapeyachtclub
Options:
--limit 500: Fetch more tokens for analysis--top 50: Show top 50 tokens--traits: Include trait distribution--rarest: Show rarest traits--algorithm [statistical|rarity_score|average|information]bashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py token pudgypenguins 1234 # port 1234 - example/test
bashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py compare azuki 1234,5678,9012 # 5678: 1234: 9012 = configured value
bashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py traits doodles
JSON:
bashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py export coolcats > rankings.json
CSV:
bashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py export coolcats --format csv > rankings.csv
bashcd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py cache --list cd ${CLAUDE_SKILL_DIR}/scripts && python3 rarity_analyzer.py cache --clear
| Algorithm | Description | Best For | |-----------|-------------|----------| | rarity_score | Sum of 1/frequency (default) | General use, matches rarity.tools | | statistical | Same as rarity_score | Backward compatibility | | average | Mean of trait rarities | Balanced scoring | | information | Entropy-based (-log2) | Information theory approach |
Works with any ERC-721/ERC-1155 collection that has:
See ${CLAUDE_SKILL_DIR}/references/errors.md for:
See ${CLAUDE_SKILL_DIR}/references/examples.md for:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,264 | 5,445 | -77% | 1 | 1 | 0% | 4,885 | 1,177 | -76% | 0 | 0 | — |
case-02 | fail→fail | 17,679 | 6,943 | -61% | 1 | 1 | 0% | 3,515 | 1,626 | -54% | 0 | 0 | — |
case-03 | fail→fail | 8,392 | 11,759 | +40% | 1 | 1 | 0% | 1,337 | 3,059 | +129% | 0 | 0 | — |
case-04 | fail→pass | 11,315 | 1,628 | -86% | 1 | 1 | 0% | 2,303 | 1,090 | -53% | 0 | 0 | — |
case-05 | fail→fail | 10,029 | 1,788 | -82% | 1 | 1 | 0% | 2,098 | 1,101 | -48% | 0 | 0 | — |
case-06 | fail→pass | 9,901 | 1,774 | -82% | 1 | 1 | 0% | 2,072 | 1,162 | -44% | 0 | 0 | — |
case-07 | fail→fail | 8,238 | 1,703 | -79% | 1 | 1 | 0% | 1,456 | 1,141 | -22% | 0 | 0 | — |
case-08 | fail→pass | 7,591 | 1,625 | -79% | 1 | 1 | 0% | 1,314 | 1,124 | -14% | 0 | 0 | — |
case-09 | fail→pass | 5,316 | 1,915 | -64% | 1 | 1 | 0% | 938 | 1,220 | +30% | 0 | 0 | — |
case-10 | fail→pass | 11,101 | 2,091 | -81% | 1 | 1 | 0% | 1,960 | 1,248 | -36% | 0 | 0 | — |
case-11 | fail→pass | 9,101 | 1,865 | -80% | 1 | 1 | 0% | 1,576 | 1,163 | -26% | 0 | 0 | — |
case-12 | pass→pass | 8,463 | 1,997 | -76% | 1 | 1 | 0% | 1,497 | 1,198 | -20% | 0 | 0 | — |
case-13 | fail→pass | 8,728 | 2,220 | -75% | 1 | 1 | 0% | 1,401 | 1,272 | -9% | 0 | 0 | — |
case-14 | pass→pass | 11,474 | 6,291 | -45% | 1 | 1 | 0% | 2,136 | 2,083 | -2% | 0 | 0 | — |
case-15 | pass→fail | 9,036 | 7,800 | -14% | 1 | 1 | 0% | 1,898 | 2,229 | +17% | 0 | 0 | — |
case-16 | pass→pass | 3,958 | 1,696 | -57% | 1 | 1 | 0% | 662 | 1,090 | +65% | 0 | 0 | — |
case-17 | fail→pass | 10,267 | 1,222 | -88% | 1 | 1 | 0% | 1,859 | 1,044 | -44% | 0 | 0 | — |
case-18 | fail→pass | 10,550 | 1,513 | -86% | 1 | 1 | 0% | 1,715 | 1,115 | -35% | 0 | 0 | — |
case-19 | pass→pass | 8,923 | 3,829 | -57% | 1 | 1 | 0% | 1,688 | 1,630 | -3% | 0 | 0 | — |
case-20 | pass→pass | 11,424 | 8,411 | -26% | 1 | 1 | 0% | 1,962 | 2,294 | +17% | 0 | 0 | — |
case-21 | pass→pass | 3,124 | 2,738 | -12% | 1 | 1 | 0% | 582 | 1,336 | +130% | 0 | 0 | — |
case-22 | pass→pass | 8,978 | 7,718 | -14% | 1 | 1 | 0% | 1,634 | 2,060 | +26% | 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 20 counted toward the lift figure. The other 2 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 +36 percentage points is the difference between those two pass rates over the 20 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.