{"id":"d194d33d-ebb6-4a7c-ab05-b3d3555771cd","shortId":"3Wky2B","kind":"skill","title":"skill-usage-log-reviewer","tagline":"Audit installed Claude skills: flag unused, spot duplicates, generate deactivation checklist. Reduces context noise from skill overload. Use when portfolio grows to 10+ skills or sessions feel slow. Triggers: 'audit my skills', 'skill usage review', 'which skills to deactivate', ","description":"# Skill Usage Log Reviewer\n\nThis skill audits your installed Claude skill collection — identifying which skills you actively use, which are dormant, and which overlap with each other. It outputs a structured audit report with a one-click deactivation checklist, helping you keep your skill portfolio lean and your sessions fast.\n\n**Input:**\n- List of installed skill names (paste, file listing, or description)\n- Optional: brief usage notes per skill (\"used daily\", \"tried once\", \"never used\")\n\n**Output:**\n- `skill-audit-report.md` — table with Keep / Review / Deactivate verdicts, reasoning, and a deactivation checklist\n\n---\n\n## Language Detection\n\nDetect the user's language from their message:\n- If Russian (or contains Cyrillic): respond in Russian\n- If English (or other Latin-script language): respond in English\n- If ambiguous: respond in the language of the trigger phrase used\n\n---\n\n## Instructions\n\n### Step 1: Collect the Skill List\n\n1. Check if Claude can see the installed skills from the conversation context (e.g., available_skills list or plugin manifest). If yes — use it directly; do not ask the user.\n2. If not available from context: ask the user to provide their skill list.\n   - Accepted formats: pasted names, `ls` output from skill folder, plugin manifest text, or free-form description\n3. Validate input is non-empty.\n   - If empty: stop and respond — \"No skills listed. Please paste your installed skill names, or describe which skills you have active in Claude.\"\n4. Check if input looks like a conversation transcript or unrelated text (no skill names detectable):\n   - If detected: stop and respond — \"This doesn't look like a skill list. Please provide the names of your installed Claude skills.\"\n\n---\n\n### Step 2: Collect Usage Context\n\n1. Check if the user already provided usage notes alongside the skill list (e.g., \"skill-x — used daily\", \"skill-y — never tried\").\n   - If usage notes are present for all skills: skip to Step 3.\n2. If usage notes are absent or partial: gather usage context with targeted questions.\n   - Group skills into batches of 5–8 and ask: \"For each skill below, tell me roughly how often you use it: daily / weekly / monthly / never.\"\n   - Do not ask per-skill individually — batch questions to reduce friction.\n3. If the user declines to answer usage questions: default all skills to \"Unknown\" and note this in the report; still run the duplicate detection step.\n\n---\n\n### Step 3: Classify Each Skill\n\nFor each skill, assign a usage tier based on answers:\n\n| Usage frequency | Tier |\n|----------------|------|\n| Daily or several times a week | **Active** |\n| Once or twice a month | **Occasional** |\n| Never used, tried once, or not sure | **Unused** |\n| Frequency unknown (user didn't answer) | **Unknown** |\n\n---\n\n### Step 4: Detect Functional Duplicates\n\n1. Compare skill names and descriptions (use the spec descriptions if visible; otherwise infer from names).\n2. Identify pairs or groups of skills with overlapping core intent:\n   - Example duplicates: `meeting-prep-briefer` + `one-to-one-prep` (both prepare for meetings)\n   - Example non-duplicates: `team-update-aggregator` + `stakeholder-adapter` (aggregation vs. formatting)\n3. For each duplicate pair:\n   - Flag the lower-value skill (e.g., less specific, less used) as a duplicate candidate.\n   - Note the overlap reason in the Reason column of the audit table.\n\n---\n\n### Step 5: Assign Audit Verdict\n\nApply this decision logic per skill:\n\n| Condition | Verdict |\n|-----------|---------|\n| Active usage | ✅ Keep |\n| Occasional usage + unique functionality | ✅ Keep |\n| Occasional usage + overlaps with another skill | ⚠️ Review |\n| Unused + unique functionality | ⚠️ Review |\n| Unused + functional duplicate | ❌ Deactivate |\n| Unused for 30+ days (confirmed) | ❌ Deactivate |\n| Unknown usage + strong duplicate detected | ⚠️ Review |\n\n---\n\n### Step 6: Generate Audit Report\n\n1. Compile results into `skill-audit-report.md` using the Output Format below.\n2. Summary stats: total skills audited, Keep / Review / Deactivate counts.\n3. Audit table: one row per skill, sorted by verdict (Deactivate first, then Review, then Keep).\n4. Deactivation checklist: list all ❌ Deactivate skills as unchecked items.\n5. Notes section: flag any collection-level observations (e.g., \"4 meeting-prep skills detected — consider consolidating to 1–2\").\n6. If file system access available (Cowork mode): save as `skill-audit-report.md` in the workspace folder and confirm path.\n7. If no file system access: display full output inline in chat.\n\n**Edge Cases:**\n- Small collection (≤5 skills): complete audit normally; add note \"Your collection is small — deactivating is optional. Focus on keeping only what you actually trigger.\"\n- All skills Active: confirm collection is healthy; output table with all Keep verdicts + \"No action needed.\"\n- Duplicate pair detected but both marked Active: set both to ✅ Keep, add observation in Notes: \"Both are active — consider which covers more of your use case long-term.\"\n\n---\n\n## Negative Cases\n\n- **Empty input:** Stop — \"No skills listed. Please paste your installed skill names, or describe which skills you have active in Claude.\"\n- **Input is a conversation log or unrelated text:** Stop — \"This doesn't look like a skill list. Please provide the names of your installed Claude skills.\"\n- **User provides plugin bundle names only (not individual skills):** Respond — \"These look like plugin names. For a detailed audit, list the individual skill names within each plugin (e.g., 'skill-spec-writer', not 'skill-builder'). You can list plugin names and I'll do a coarser audit.\"\n\n---\n\n## Output Format\n\n```markdown\n# Skill Audit Report\n**Date:** YYYY-MM-DD\n**Total skills audited:** N\n**Keep:** X | **Review:** Y | **Deactivate:** Z\n\n---\n\n## Summary\n\n[1–2 sentences: overall collection health and main recommended action]\n\n---\n\n## Audit Table\n\n| Skill | Verdict | Reason |\n|-------|---------|--------|\n| skill-name | ❌ Deactivate | Not used in 30+ days; duplicate of skill-y |\n| skill-name | ⚠️ Review | Used once; overlaps with skill-z |\n| skill-name | ✅ Keep | Used weekly for meeting prep |\n\n---\n\n## Deactivation Checklist\n\nSkills recommended for deactivation — go to Claude settings and remove each:\n\n- [ ] skill-name — reason: not used in 30+ days; duplicate of skill-y\n- [ ] skill-name — reason: never triggered; functionality covered by skill-z\n\n*(If no deactivations: \"No skills recommended for deactivation.\")*\n\n---\n\n## Notes\n\n[Collection-level observations, e.g., \"3 meeting-prep skills detected — consider consolidating\"]\n```\n\n**Field rules:**\n- Verdict uses emoji prefix: ✅ Keep / ⚠️ Review / ❌ Deactivate\n- Reason column: ≤15 words, specific (not generic like \"not needed\")\n- Audit table sorted: ❌ first, then ⚠️, then ✅\n- Deactivation Checklist: only ❌ Deactivate verdicts appear here\n- Notes section: always present; \"No additional observations.\" if nothing to flag","tags":["skill","usage","log","reviewer","claude","skills","kit","kirkruglov","agent-skills","agentic-skills","ai-agents","ai-skills"],"capabilities":["skill","source-kirkruglov","skill-skill-usage-log-reviewer","topic-agent-skills","topic-agentic-skills","topic-ai-agents","topic-ai-skills","topic-awesome-list","topic-claude","topic-claude-ai","topic-claude-ai-skills","topic-claude-code","topic-claude-cowork","topic-claude-memory","topic-claude-skills"],"categories":["claude-skills-kit"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/KirKruglov/claude-skills-kit/skill-usage-log-reviewer","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add KirKruglov/claude-skills-kit","source_repo":"https://github.com/KirKruglov/claude-skills-kit","install_from":"skills.sh"}},"qualityScore":"0.453","qualityRationale":"deterministic score 0.45 from registry signals: · indexed on github topic:agent-skills · 7 github stars · SKILL.md body (6,927 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T19:13:38.971Z","embedding":null,"createdAt":"2026-05-18T13:21:16.912Z","updatedAt":"2026-05-18T19:13:38.971Z","lastSeenAt":"2026-05-18T19:13:38.971Z","tsv":"'1':174,179,313,477,618,683,905 '10':28 '15':1026 '2':209,309,349,493,628,684,906 '3':240,348,400,427,533,638,1007 '30':603,927,974 '4':270,473,654,674 '5':368,566,664,719 '6':614,685 '7':703 '8':369 'absent':354 'accept':223 'access':689,708 'action':755,914 'activ':61,267,450,578,743,763,774,806 'actual':739 'adapt':529 'add':724,768 'addit':1052 'aggreg':526,530 'alongsid':322 'alreadi':318 'alway':1049 'ambigu':162 'anoth':590 'answer':406,440,470 'appear':1045 'appli':570 'ask':206,215,371,390 'assign':434,567 'audit':6,35,51,76,563,568,616,633,639,722,853,882,887,896,915,1034 'avail':193,212,690 'base':438 'batch':366,395 'brief':108 'briefer':509 'builder':870 'bundl':838 'candid':552 'case':716,782,787 'chat':714 'check':180,271,314 'checklist':16,84,131,656,955,1041 'classifi':428 'claud':8,54,182,269,306,808,833,962 'click':82 'coarser':881 'collect':56,175,310,670,718,727,745,909,1003 'collection-level':669,1002 'column':560,1025 'compar':478 'compil':619 'complet':721 'condit':576 'confirm':605,701,744 'consid':680,775,1013 'consolid':681,1014 'contain':145 'context':18,191,214,312,359 'convers':190,277,812 'core':502 'count':637 'cover':777,988 'cowork':691 'cyril':146 'daili':114,331,384,444 'date':889 'day':604,928,975 'dd':893 'deactiv':15,44,83,125,130,600,606,636,648,655,659,730,902,923,954,959,995,1000,1023,1040,1043 'decis':572 'declin':404 'default':409 'describ':262,801 'descript':106,239,482,486 'detail':852 'detect':133,134,285,287,424,474,611,679,759,1012 'didn':468 'direct':203 'display':709 'doesn':292,819 'dormant':65 'duplic':13,423,476,505,522,536,551,599,610,757,929,976 'e.g':192,326,544,673,862,1006 'edg':715 'emoji':1019 'empti':246,248,788 'english':151,160 'exampl':504,519 'fast':95 'feel':32 'field':1015 'file':103,687,706 'first':649,1037 'flag':10,538,667,1057 'focus':733 'folder':231,699 'form':238 'format':224,532,626,884 'free':237 'free-form':236 'frequenc':442,465 'friction':399 'full':710 'function':475,584,595,598,987 'gather':357 'generat':14,615 'generic':1030 'go':960 'group':363,497 'grow':26 'health':910 'healthi':747 'help':85 'identifi':57,494 'individu':394,842,856 'infer':490 'inlin':712 'input':96,242,273,789,809 'instal':7,53,99,186,258,305,797,832 'instruct':172 'intent':503 'item':663 'keep':87,123,580,585,634,653,735,752,767,898,948,1021 'languag':132,138,157,166 'latin':155 'latin-script':154 'lean':91 'less':545,547 'level':671,1004 'like':275,295,822,847,1031 'list':97,104,178,195,222,254,298,325,657,793,825,854,873 'll':878 'log':4,47,813 'logic':573 'long':784 'long-term':783 'look':274,294,821,846 'lower':541 'lower-valu':540 'ls':227 'main':912 'manifest':198,233 'mark':762 'markdown':885 'meet':507,518,676,952,1009 'meeting-prep':675,1008 'meeting-prep-brief':506 'messag':141 'mm':892 'mode':692 'month':386,455 'n':897 'name':101,226,260,284,302,480,492,799,829,839,849,858,875,922,936,947,969,983 'need':756,1033 'negat':786 'never':117,335,387,457,985 'nois':19 'non':245,521 'non-dupl':520 'non-empti':244 'normal':723 'note':110,321,339,352,415,553,665,725,771,1001,1047 'noth':1055 'observ':672,769,1005,1053 'occasion':456,581,586 'often':380 'one':81,511,513,641 'one-click':80 'one-to-one-prep':510 'option':107,732 'otherwis':489 'output':73,119,228,625,711,748,883 'overal':908 'overlap':68,501,555,588,940 'overload':22 'pair':495,537,758 'partial':356 'past':102,225,256,795 'path':702 'per':111,392,574,643 'per-skil':391 'phrase':170 'pleas':255,299,794,826 'plugin':197,232,837,848,861,874 'portfolio':25,90 'prefix':1020 'prep':508,514,677,953,1010 'prepar':516 'present':341,1050 'provid':219,300,319,827,836 'question':362,396,408 'reason':127,556,559,919,970,984,1024 'recommend':913,957,998 'reduc':17,398 'remov':965 'report':77,419,617,888 'respond':147,158,163,251,290,844 'result':620 'review':5,40,48,124,592,596,612,635,651,900,937,1022 'rough':378 'row':642 'rule':1016 'run':421 'russian':143,149 'save':693 'script':156 'section':666,1048 'see':184 'sentenc':907 'session':31,94 'set':764,963 'sever':446 'skill':2,9,21,29,37,38,42,45,50,55,59,89,100,112,177,187,194,221,230,253,259,264,283,297,307,324,328,333,344,364,374,393,411,430,433,479,499,543,575,591,632,644,660,678,720,742,792,798,803,824,834,843,857,864,869,886,895,917,921,932,935,943,946,956,968,979,982,991,997,1011 'skill-audit-report.md':120,622,695 'skill-build':868 'skill-i':332,931,978 'skill-nam':920,934,945,967,981 'skill-skill-usage-log-reviewer' 'skill-spec-writ':863 'skill-usage-log-review':1 'skill-x':327 'skill-z':942,990 'skip':345 'slow':33 'small':717,729 'sort':645,1036 'source-kirkruglov' 'spec':485,865 'specif':546,1028 'spot':12 'stakehold':528 'stakeholder-adapt':527 'stat':630 'step':173,308,347,425,426,472,565,613 'still':420 'stop':249,288,790,817 'strong':609 'structur':75 'summari':629,904 'sure':463 'system':688,707 'tabl':121,564,640,749,916,1035 'target':361 'team':524 'team-update-aggreg':523 'tell':376 'term':785 'text':234,281,816 'tier':437,443 'time':447 'topic-agent-skills' 'topic-agentic-skills' 'topic-ai-agents' 'topic-ai-skills' 'topic-awesome-list' 'topic-claude' 'topic-claude-ai' 'topic-claude-ai-skills' 'topic-claude-code' 'topic-claude-cowork' 'topic-claude-memory' 'topic-claude-skills' 'total':631,894 'transcript':278 'tri':115,336,459 'trigger':34,169,740,986 'twice':453 'uncheck':662 'uniqu':583,594 'unknown':413,466,471,607 'unrel':280,815 'unus':11,464,593,597,601 'updat':525 'usag':3,39,46,109,311,320,338,351,358,407,436,441,579,582,587,608 'use':23,62,113,118,171,201,330,382,458,483,548,623,781,925,938,949,972,1018 'user':136,208,217,317,403,467,835 'valid':241 'valu':542 'verdict':126,569,577,647,753,918,1017,1044 'visibl':488 'vs':531 'week':385,449,950 'within':859 'word':1027 'workspac':698 'writer':866 'x':329,899 'y':334,901,933,980 'yes':200 'yyyi':891 'yyyy-mm-dd':890 'z':903,944,992","prices":[{"id":"1bd65d01-4a88-4f25-9640-98b3ddc72a61","listingId":"d194d33d-ebb6-4a7c-ab05-b3d3555771cd","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"KirKruglov","category":"claude-skills-kit","install_from":"skills.sh"},"createdAt":"2026-05-18T13:21:16.912Z"}],"sources":[{"listingId":"d194d33d-ebb6-4a7c-ab05-b3d3555771cd","source":"github","sourceId":"KirKruglov/claude-skills-kit/skill-usage-log-reviewer","sourceUrl":"https://github.com/KirKruglov/claude-skills-kit/tree/main/skills/skill-usage-log-reviewer","isPrimary":false,"firstSeenAt":"2026-05-18T13:21:16.912Z","lastSeenAt":"2026-05-18T19:13:38.971Z"}],"details":{"listingId":"d194d33d-ebb6-4a7c-ab05-b3d3555771cd","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"KirKruglov","slug":"skill-usage-log-reviewer","github":{"repo":"KirKruglov/claude-skills-kit","stars":7,"topics":["agent-skills","agentic-skills","ai","ai-agents","ai-skills","awesome-list","claude","claude-ai","claude-ai-skills","claude-code","claude-cowork","claude-memory","claude-skills","memory-management","productivity","productivity-tools","project-management"],"license":"mit","html_url":"https://github.com/KirKruglov/claude-skills-kit","pushed_at":"2026-05-18T04:27:46Z","description":"40+ curated agent skills for Claude Cowork and Claude.ai — ready-to-use tools for non-technical users: project management, productivity, and AI workflow automation","skill_md_sha":"9c4ca6ffb49c974d4473277a6ac139bca08a2f20","skill_md_path":"skills/skill-usage-log-reviewer/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/KirKruglov/claude-skills-kit/tree/main/skills/skill-usage-log-reviewer"},"layout":"multi","source":"github","category":"claude-skills-kit","frontmatter":{"name":"skill-usage-log-reviewer","description":"Audit installed Claude skills: flag unused, spot duplicates, generate deactivation checklist. Reduces context noise from skill overload. Use when portfolio grows to 10+ skills or sessions feel slow. Triggers: 'audit my skills', 'skill usage review', 'which skills to deactivate', 'аудит скилов', 'какие скилы деактивировать'."},"skills_sh_url":"https://skills.sh/KirKruglov/claude-skills-kit/skill-usage-log-reviewer"},"updatedAt":"2026-05-18T19:13:38.971Z"}}