{"id":"88d77435-25ec-4bd4-ae09-c9274d63c10d","shortId":"LxyJs8","kind":"skill","title":"skill-optimizer","tagline":"Use when the user wants to analyze, audit, or improve their Agent Skills (SKILL.md files). Triggers on /optimize-skill, /skill-audit, 'optimize skills', 'analyze skills', 'check my skills', 'skill quality'. Also use proactively when the user mentions skills aren't triggering, ski","description":"## Rules\n\n- **Read-only**: never modify skill files. Only output report.\n- **All 8 dimensions**: do not skip any. If data is insufficient, report \"N/A — insufficient session data\" rather than omitting.\n- **Quantify**: \"you had 12 research tasks last week but the skill never triggered\" beats \"you often do research\".\n- **Suggest, don't prescribe**: give specific wording suggestions for description improvements, but frame as suggestions.\n- **Show evidence**: for undertrigger claims, quote the actual user message that should have triggered the skill.\n- **Evidence-based suggestions**: when suggesting description rewrites, cite the specific research finding that motivates the change (e.g., \"front-load trigger keywords — MCP study shows 3.6x selection rate improvement\").\n\n## Overview\n\nAnalyze skills using **historical session data + static quality checks**, output a diagnostic report with P0/P1/P2 prioritized fixes. Scores each skill on a 5-point composite scale across 8 dimensions.\n\nCSO (Claude/Agent Search Optimization) = writing skill descriptions so agents select the right skill at the right time. This skill checks for CSO violations.\n\n## Usage\n\n- `/optimize-skill` → scan all skills\n- `/optimize-skill my-skill` → single skill\n- `/optimize-skill skill-a skill-b` → multiple specified skills\n\n## Data Sources\n\nAuto-detect the current agent platform and scan the corresponding paths:\n\n| Source | Claude Code | Codex | Shared |\n|--------|------------|-------|--------|\n| Session transcripts | `~/.claude/projects/**/*.jsonl` | `~/.codex/sessions/**/*.jsonl` | — |\n| Skill files | `~/.claude/skills/*/SKILL.md` | `~/.codex/skills/*/SKILL.md` | `~/.agents/skills/*/SKILL.md` |\n\n**Platform detection:** Check which directories exist. Scan all available sources — a user may have both Claude Code and Codex installed.\n\n## Workflow\n\n```\nIdentify target skills\n        ↓\nCollect session data (python3 scripts scan JSONL transcripts)\n        ↓\nRun 8 analysis dimensions\n        ↓\nCompute composite scores\n        ↓\nOutput report with P0/P1/P2\n```\n\n### Step 1: Identify Target Skills\n\nScan skill directories in order: `~/.claude/skills/`, `~/.codex/skills/`, `~/.agents/skills/`. Deduplicate by skill name (same name in multiple locations = same skill). For each, read `SKILL.md` and extract:\n- name, description (from YAML frontmatter)\n- trigger keywords (from description field)\n- defined workflow steps (Step 1/2/3... or ### sections under Workflow)\n- word count\n\nIf user specified skill names, filter to only those.\n\n### Step 2: Collect Session Data\n\nUse python3 scripts via Bash to scan session JSONL files. Extract:\n\n**Claude Code sessions** (`~/.claude/projects/**/*.jsonl`):\n- `Skill` tool_use calls (which skills were invoked)\n- User messages (full text)\n- Assistant messages after skill invocation (for workflow tracking)\n- User messages after skill invocation (for reaction analysis)\n\n**Codex sessions** (`~/.codex/sessions/**/*.jsonl`):\n- `session_meta` events → extract `base_instructions` for skill loading evidence\n- `response_item` events → assistant outputs (workflow tracking)\n- `event_msg` events → tool execution and skill-related events\n- User messages from `turn_context` events (for reaction analysis)\n\n**Note:** Codex injects skills via context rather than explicit `Skill` tool calls. Skill loading (present in `base_instructions`) does NOT equal active invocation. To detect actual use, search for skill-specific workflow markers (step headers, output formats) in `response_item` content within that session. A skill is \"invoked\" only if the agent produced output following the skill's defined workflow.\n\n**Aggregated:**\n- Per-skill: invocation count, trigger keyword match count\n- Per-skill: user reaction sentiment after invocation\n- Per-skill: workflow step completion markers\n\n### Step 3: Run 8 Analysis Dimensions\n\n**You MUST run ALL 8 dimensions.** The baseline behavior without this skill is to skip dimensions 4.2, 4.3, 4.5b, and 4.8. These are the most valuable dimensions — do not skip them.\n\n#### 4.1 Trigger Rate\n\nCount how many times each skill was actually invoked vs how many times its trigger keywords appeared in user messages.\n\n**Claude Code:** count `Skill` tool_use calls in transcripts.\n**Codex:** count sessions where the agent produced output following the skill's workflow markers (not merely loaded in context).\n\n**Diagnose:**\n- Never triggered → skill may be useless or trigger words wrong\n- Keywords match >> actual invocations → undertrigger problem, description needs work\n- High frequency → core skill, worth optimizing\n\n#### 4.2 Post-Invocation User Reaction\n\n**This dimension is critical and easy to skip. Do not skip it.**\n\nAfter a skill is invoked in a session, read the user's next 3 messages. Classify:\n- **Negative**: \"no\", \"wrong\", \"never mind\", \"not what I wanted\", user interrupts\n- **Correction**: user re-describes their intent, manually overrides skill output\n- **Positive**: \"good\", \"ok\", \"continue\", \"nice\", user follows the workflow\n- **Silent switch**: user changes topic entirely (likely false positive trigger)\n\nReport per-skill satisfaction rate.\n\n#### 4.3 Workflow Completion Rate\n\n**This dimension is critical and easy to skip. Do not skip it.**\n\nFor each skill invocation found in session data:\n1. Extract the skill's defined steps from SKILL.md\n2. Search the assistant messages in that session for step markers (Step N, specific output formats defined in the skill)\n3. Calculate: how far did execution get?\n\nReport: `{skill-name} (N steps): avg completed Step X/N (Y%)`\n\nIf a specific step is frequently where execution stops, flag it.\n\n#### 4.4 Static Quality Analysis\n\nCheck each SKILL.md against these 14 rules:\n\n| Check | Pass Criteria |\n|-------|--------------|\n| Frontmatter format | Only `name` + `description`, total < 1024 chars |\n| Name format | Letters, numbers, hyphens only |\n| Description trigger | Starts with \"Use when...\" or has explicit trigger conditions |\n| Description workflow leak | Description does NOT summarize the skill's workflow steps (CSO violation) |\n| Description pushiness | Description actively claims scenarios where it should be used, not just passive |\n| Overview section | Present |\n| Rules section | Present |\n| MUST/NEVER density | Count ALL-CAPS directive words; >5 per 100 words = flag. Note: Meincke et al. (2025) found persuasion directives have inconsistent effects across models. Suggest converting to concrete bright-line rules with rationale, not mere emphasis. |\n| Word count | < 500 words (flag if over) |\n| Narrative anti-pattern | No \"In session X, we found...\" storytelling — skills should be instructions, not post-hoc reports |\n| YAML quoting safety | description containing `: ` must be wrapped in double quotes, otherwise YAML parse failure makes skill invisible |\n| Critical info position | Core trigger conditions and primary actions must be in the first 20% of SKILL.md, not buried in the middle (Lost in the Middle, Liu et al. TACL 2024: U-shaped attention curve) |\n| Description 250-char check | Primary trigger keywords must appear within the first 250 characters of description (skill listing truncation point in most agents) |\n| Trigger condition count | ≤ 2 trigger conditions in description is ideal; consistent with IFEval (Zhou et al. 2023) finding that LLMs struggle with multi-constraint prompts |\n\n#### 4.5a False Positive Rate (Overtrigger)\n\nSkill was invoked but user immediately rejected or ignored it.\n\n#### 4.5b Undertrigger Detection\n\n**This is the highest-value dimension.** Memento-Skills (arXiv:2603.18743) demonstrates that skills stored as structured files require accurate retrieval/routing to be effective — skills that are never retrieved cannot improve through their read-write learning loop, making undertriggering a compounding problem.\n\nFor each skill, extract its **capability keywords** (not just trigger keywords — what the skill CAN do). Then scan user messages for tasks that match those capabilities but where the skill was NOT invoked.\n\nExample: user says \"run these tasks in parallel\" but parallel-runner was not triggered → undertrigger.\n\nReport: which user messages SHOULD have triggered the skill but didn't, and suggest description improvements.\n\n**Compounding Risk Assessment:**\nFor skills with chronic undertriggering (0 triggers across 5+ sessions where relevant tasks appeared), flag as \"compounding risk\" — undertriggered skills cannot self-improve through usage feedback, causing the gap to widen over time. Recommend immediate description rewrite as P0.\n\n#### 4.6 Cross-Skill Conflicts\n\nCompare all skill pairs:\n- Trigger keyword overlap (same keywords in two descriptions)\n- Workflow overlap (two skills teach similar processes)\n- Contradictory guidance\n\n#### 4.7 Environment Consistency\n\nFor each skill, extract referenced:\n- File paths → check if they exist (`test -e`)\n- CLI tools → check if installed (`which`)\n- Directories → check if they exist\n\nFlag any broken references.\n\n#### 4.8 Token Economics\n\n**This dimension is critical and easy to skip. Do not skip it.**\n\nFor each skill:\n- Word count (from Step 1)\n- Trigger frequency (from 4.1)\n- Cost-effectiveness = trigger count / word count\n- Flag: large + never-triggered skills as candidates for removal or compression\n\n**Progressive Disclosure Tier Check:**\nEvaluate each skill against the 3-tier loading model (Agent Skills spec):\n- Tier 1 (frontmatter): ~100 tokens. Check: is description ≤ 1024 chars?\n- Tier 2 (SKILL.md body): <500 lines recommended. Check: word count.\n- Tier 3 (reference files): loaded on demand. Check: does skill use reference files for detailed content, or cram everything into SKILL.md?\n\nFlag skills that put 500+ words in SKILL.md without using reference files as \"poor progressive disclosure\".\n\n### Step 4: Composite Score\n\nRate each skill on a 5-point scale:\n\n| Score | Meaning |\n|-------|---------|\n| 5 | Healthy: high trigger rate, positive reactions, complete workflows, clean static |\n| 4 | Good: minor issues in 1-2 dimensions |\n| 3 | Needs attention: significant gap in 1 dimension or minor gaps in 3+ |\n| 2 | Problematic: never triggered, or negative user reactions, or major static issues |\n| 1 | Broken: doesn't work, references missing, or fundamentally misaligned |\n\n**Scored dimensions** (weighted average):\n- Trigger rate: 25%\n- User reaction: 20%\n- Workflow completion: 15%\n- Static quality: 15%\n- Undertrigger: 15%\n- Token economics: 10%\n\n**Qualitative dimensions** (reported but not scored — no reliable numeric metric):\n- 4.5a Overtrigger: reported as count + examples\n- 4.6 Cross-Skill Conflicts: reported as conflict pairs\n- 4.7 Environment Consistency: reported as pass/fail per reference\n\n(If a scored dimension has no data — e.g., skill was never invoked so no user reaction — mark as \"N/A\" and redistribute weight.)\n\n## Report Format\n\n```markdown\n# Skill Optimization Report\n**Date**: {date}\n**Scope**: {all / specified skills}\n**Session data**: {N} sessions, {date range}\n\n## Overview\n| Skill | Triggers | Reaction | Completion | Static | Undertrigger | Token | Score |\n|-------|----------|----------|------------|--------|--------------|-------|-------|\n| example-skill | 2 | 100% | 86% | B+ | 1 miss | 486w | 4/5 |\n\n## P0 Fixes (blocking usage)\n1. ...\n\n## P1 Improvements (better experience)\n1. ...\n\n## P2 Optional Optimizations\n1. ...\n\n## Per-Skill Diagnostics\n### {skill-name}\n#### 4.1 Trigger Rate\n...\n#### 4.2 User Reaction\n...\n(all 8 dimensions)\n```\n\n## Research Background\nThe analysis dimensions in this report are grounded in the following research:\n- **Undertrigger detection**: Memento-Skills (arXiv:2603.18743) — skills as structured files require accurate routing; unrouted skills cannot self-improve via the read-write learning loop\n- **Description quality**: MCP Description Quality (arXiv:2602.18914) — well-written descriptions achieve 72% tool selection rate vs. 20% random baseline (3.6x improvement)\n- **Information position**: Lost in the Middle (Liu et al., TACL 2024) — U-shaped LLM attention curve\n- **Format impact**: He et al. (arXiv:2411.10541) — format changes alone can cause 9-40% performance variance\n- **Instruction compliance**: IFEval (arXiv:2311.07911) — LLMs struggle with multi-constraint prompts","tags":["skill","optimizer","hqhq1025","agent-skills","agentskills","claude-code","claude-skill","codex","llm-agents","prompt-engineering","skill-optimizer"],"capabilities":["skill","source-hqhq1025","skill-skill-optimizer","topic-agent-skills","topic-agentskills","topic-claude-code","topic-claude-skill","topic-codex","topic-llm-agents","topic-prompt-engineering","topic-skill-optimizer"],"categories":["skill-optimizer"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/hqhq1025/skill-optimizer/skill-optimizer","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add hqhq1025/skill-optimizer","source_repo":"https://github.com/hqhq1025/skill-optimizer","install_from":"skills.sh"}},"qualityScore":"0.483","qualityRationale":"deterministic score 0.48 from registry signals: · indexed on github topic:agent-skills · 67 github stars · SKILL.md body (12,242 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":"2026-05-02T06:55:48.668Z","embedding":null,"createdAt":"2026-04-18T22:16:10.683Z","updatedAt":"2026-05-02T06:55:48.668Z","lastSeenAt":"2026-05-02T06:55:48.668Z","tsv":"'-2':1438 '-40':1714 '/.agents/skills':259,316 '/.claude/projects':249,383 '/.claude/skills':255,314 '/.codex/sessions':251,415 '/.codex/skills':257,315 '/optimize-skill':21,208,212,218 '/skill-audit':22 '/skill.md':256,258,260 '0':1196 '1':305,759,1310,1351,1437,1446,1465,1586,1594,1599,1603 '1/2/3':348 '10':1495 '100':900,1353,1583 '1024':837,1358 '12':77 '14':826 '15':1487,1490,1492 '2':365,768,1036,1361,1453,1582 '20':988,1484,1678 '2023':1049 '2024':1004,1694 '2025':907 '2311.07911':1721 '2411.10541':1707 '25':1481 '250':1011,1022 '2602.18914':1667 '2603.18743':1090,1640 '3':540,685,788,1343,1371,1440,1452 '3.6':149,1681 '4':1408,1432 '4.1':577,1314,1611 '4.2':561,654,1614 '4.3':562,735 '4.4':817 '4.5':563,1059,1075,1506 '4.6':1231,1513 '4.7':1257,1522 '4.8':566,1288 '4/5':1589 '486w':1588 '5':177,898,1199,1416,1421 '500':931,1364,1395 '72':1673 '8':56,182,294,542,549,1618 '86':1584 '9':1713 'accur':1099,1646 'achiev':1672 'across':181,914,1198 'action':982 'activ':474,873 'actual':114,478,587,641 'agent':15,192,235,505,614,1032,1347 'aggreg':514 'al':906,1002,1048,1692,1705 'all-cap':893 'alon':1710 'also':32 'analysi':295,412,452,543,820,1623 'analyz':10,25,155 'anti':938 'anti-pattern':937 'appear':596,1018,1204 'aren':40 'arxiv':1089,1639,1666,1706,1720 'assess':1190 'assist':397,430,771 'attent':1008,1442,1699 'audit':11 'auto':231 'auto-detect':230 'avail':269 'averag':1478 'avg':801 'b':224,564,1076,1585 'background':1621 'base':125,421,469 'baselin':552,1680 'bash':373 'beat':87 'behavior':553 'better':1597 'block':1592 'bodi':1363 'bright':921 'bright-lin':920 'broken':1286,1466 'buri':992 'calcul':789 'call':388,464,606 'candid':1329 'cannot':1109,1211,1650 'cap':895 'capabl':1128,1148 'caus':1218,1712 'chang':139,722,1709 'char':838,1012,1359 'charact':1023 'check':27,163,203,263,821,828,1013,1267,1275,1280,1337,1355,1367,1377 'chronic':1194 'cite':131 'claim':111,874 'classifi':687 'claud':243,276,380,600 'claude/agent':185 'clean':1430 'cli':1273 'code':244,277,381,601 'codex':245,279,413,454,609 'collect':285,366 'compar':1236 'complet':537,737,802,1428,1486,1574 'complianc':1718 'composit':179,298,1409 'compound':1121,1188,1207 'compress':1333 'comput':297 'concret':919 'condit':855,979,1034,1038 'conflict':1235,1517,1520 'consist':1043,1259,1524 'constraint':1057,1727 'contain':960 'content':494,1385 'context':448,458,627 'continu':713 'contradictori':1255 'convert':917 'core':650,977 'correct':699 'correspond':240 'cost':1316 'cost-effect':1315 'count':354,519,523,580,602,610,892,930,1035,1307,1319,1321,1369,1511 'cram':1387 'criteria':830 'critic':663,742,974,1294 'cross':1233,1515 'cross-skil':1232,1514 'cso':184,205,868 'current':234 'curv':1009,1700 'data':63,70,160,228,287,368,758,1536,1565 'date':1558,1559,1568 'dedupl':317 'defin':344,512,764,784 'demand':1376 'demonstr':1091 'densiti':891 'describ':703 'descript':101,129,190,335,342,645,835,845,856,859,870,872,959,1010,1025,1040,1186,1227,1247,1357,1661,1664,1671 'detail':1384 'detect':232,262,477,1078,1635 'diagnos':628 'diagnost':166,1607 'didn':1182 'dimens':57,183,296,544,550,560,572,661,740,1085,1292,1439,1447,1476,1497,1533,1619,1624 'direct':896,910 'directori':265,311,1279 'disclosur':1335,1406 'doesn':1467 'doubl':965 'e':1272 'e.g':140,1537 'easi':665,744,1296 'econom':1290,1494 'effect':913,1103,1317 'emphasi':928 'entir':724 'environ':1258,1523 'equal':473 'et':905,1001,1047,1691,1704 'evalu':1338 'event':419,429,434,436,443,449 'everyth':1388 'evid':108,124,426 'evidence-bas':123 'exampl':1156,1512,1580 'example-skil':1579 'execut':438,793,813 'exist':266,1270,1283 'experi':1598 'explicit':461,853 'extract':333,379,420,760,1126,1263 'failur':970 'fals':726,1061 'far':791 'feedback':1217 'field':343 'file':18,51,254,378,1097,1265,1373,1382,1402,1644 'filter':360 'find':135,1050 'first':987,1021 'fix':171,1591 'flag':815,902,933,1205,1284,1322,1391 'follow':508,617,716,1632 'format':490,783,832,840,1553,1701,1708 'found':755,908,945 'frame':104 'frequenc':649,1312 'frequent':811 'front':142 'front-load':141 'frontmatt':338,831,1352 'full':395 'fundament':1473 'gap':1220,1444,1450 'get':794 'give':96 'good':711,1433 'ground':1629 'guidanc':1256 'header':488 'healthi':1422 'high':648,1423 'highest':1083 'highest-valu':1082 'histor':158 'hoc':954 'hyphen':843 'ideal':1042 'identifi':282,306 'ifev':1045,1719 'ignor':1073 'immedi':1070,1226 'impact':1702 'improv':13,102,153,1110,1187,1214,1596,1653,1683 'inconsist':912 'info':975 'inform':1684 'inject':455 'instal':280,1277 'instruct':422,470,950,1717 'insuffici':65,68 'intent':705 'interrupt':698 'invis':973 'invoc':401,409,475,518,531,642,657,754 'invok':392,501,588,676,1067,1155,1541 'issu':1435,1464 'item':428,493 'jsonl':250,252,291,377,384,416 'keyword':145,340,521,595,639,1016,1129,1133,1241,1244 'larg':1323 'last':80 'leak':858 'learn':1116,1659 'letter':841 'like':725 'line':922,1365 'list':1027 'liu':1000,1690 'llm':1698 'llms':1052,1722 'load':143,425,466,625,1345,1374 'locat':325 'loop':1117,1660 'lost':996,1686 'major':1462 'make':971,1118 'mani':582,591 'manual':706 'mark':1546 'markdown':1554 'marker':486,538,622,778 'match':522,640,1146 'may':273,632 'mcp':146,1663 'mean':1420 'meinck':904 'memento':1087,1637 'memento-skil':1086,1636 'mention':38 'mere':624,927 'messag':116,394,398,406,445,599,686,772,1142,1175 'meta':418 'metric':1505 'middl':995,999,1689 'mind':692 'minor':1434,1449 'misalign':1474 'miss':1471,1587 'model':915,1346 'modifi':49 'motiv':137 'msg':435 'multi':1056,1726 'multi-constraint':1055,1725 'multipl':225,324 'must':546,961,983,1017 'must/never':890 'my-skil':213 'n':780,799,1566 'n/a':67,1548 'name':320,322,334,359,798,834,839,1610 'narrat':936 'need':646,1441 'negat':688,1458 'never':48,85,629,691,1107,1325,1455,1540 'never-trigg':1324 'next':684 'nice':714 'note':453,903 'number':842 'numer':1504 'often':89 'ok':712 'omit':73 'optim':3,23,187,653,1556,1602 'option':1601 'order':313 'otherwis':967 'output':53,164,300,431,489,507,616,709,782 'overlap':1242,1249 'overrid':707 'overtrigg':1064,1508 'overview':154,884,1570 'p0':1230,1590 'p0/p1/p2':169,303 'p1':1595 'p2':1600 'pair':1239,1521 'parallel':1163,1166 'parallel-runn':1165 'pars':969 'pass':829 'pass/fail':1527 'passiv':883 'path':241,1266 'pattern':939 'per':516,525,533,731,899,1528,1605 'per-skil':515,524,532,730,1604 'perform':1715 'persuas':909 'platform':236,261 'point':178,1029,1417 'poor':1404 'posit':710,727,976,1062,1426,1685 'post':656,953 'post-hoc':952 'post-invoc':655 'prescrib':95 'present':467,886,889 'primari':981,1014 'priorit':170 'proactiv':34 'problem':644,1122 'problemat':1454 'process':1254 'produc':506,615 'progress':1334,1405 'prompt':1058,1728 'pushi':871 'put':1394 'python3':288,370 'qualit':1496 'qualiti':31,162,819,1489,1662,1665 'quantifi':74 'quot':112,957,966 'random':1679 'rang':1569 'rate':152,579,734,738,1063,1411,1425,1480,1613,1676 'rather':71,459 'rational':925 're':702 're-describ':701 'reaction':411,451,528,659,1427,1460,1483,1545,1573,1616 'read':46,330,680,1114,1657 'read-on':45 'read-writ':1113,1656 'recommend':1225,1366 'redistribut':1550 'refer':1287,1372,1381,1401,1470,1529 'referenc':1264 'reject':1071 'relat':442 'relev':1202 'reliabl':1503 'remov':1331 'report':54,66,167,301,729,795,955,1172,1498,1509,1518,1525,1552,1557,1627 'requir':1098,1645 'research':78,91,134,1620,1633 'respons':427,492 'retriev':1108 'retrieval/routing':1100 'rewrit':130,1228 'right':195,199 'risk':1189,1208 'rout':1647 'rule':44,827,887,923 'run':293,541,547,1159 'runner':1167 'safeti':958 'satisfact':733 'say':1158 'scale':180,1418 'scan':209,238,267,290,309,375,1140 'scenario':875 'scope':1560 'score':172,299,1410,1419,1475,1501,1532,1578 'script':289,371 'search':186,480,769 'section':350,885,888 'select':151,193,1675 'self':1213,1652 'self-improv':1212,1651 'sentiment':529 'session':69,159,247,286,367,376,382,414,417,497,611,679,757,775,942,1200,1564,1567 'shape':1007,1697 'share':246 'show':107,148 'signific':1443 'silent':719 'similar':1253 'singl':216 'ski':43 'skill':2,16,24,26,29,30,39,50,84,122,156,174,189,196,202,211,215,217,220,223,227,253,284,308,310,319,327,358,385,390,400,408,424,441,456,462,465,483,499,510,517,526,534,556,585,603,619,631,651,674,708,732,753,762,787,797,864,947,972,1026,1065,1088,1093,1104,1125,1136,1152,1180,1192,1210,1234,1238,1251,1262,1305,1327,1340,1348,1379,1392,1413,1516,1538,1555,1563,1571,1581,1606,1609,1638,1641,1649 'skill-a':219 'skill-b':222 'skill-nam':796,1608 'skill-optim':1 'skill-rel':440 'skill-skill-optimizer' 'skill-specif':482 'skill.md':17,331,767,823,990,1362,1390,1398 'skip':60,559,575,667,670,746,749,1298,1301 'sourc':229,242,270 'source-hqhq1025' 'spec':1349 'specif':97,133,484,781,808 'specifi':226,357,1562 'start':847 'static':161,818,1431,1463,1488,1575 'step':304,346,347,364,487,536,539,765,777,779,800,803,809,867,1309,1407 'stop':814 'store':1094 'storytel':946 'structur':1096,1643 'struggl':1053,1723 'studi':147 'suggest':92,99,106,126,128,916,1185 'summar':862 'switch':720 'tacl':1003,1693 'target':283,307 'task':79,1144,1161,1203 'teach':1252 'test':1271 'text':396 'tier':1336,1344,1350,1360,1370 'time':200,583,592,1224 'token':1289,1354,1493,1577 'tool':386,437,463,604,1274,1674 'topic':723 'topic-agent-skills' 'topic-agentskills' 'topic-claude-code' 'topic-claude-skill' 'topic-codex' 'topic-llm-agents' 'topic-prompt-engineering' 'topic-skill-optimizer' 'total':836 'track':404,433 'transcript':248,292,608 'trigger':19,42,86,120,144,339,520,578,594,630,636,728,846,854,978,1015,1033,1037,1132,1170,1178,1197,1240,1311,1318,1326,1424,1456,1479,1572,1612 'truncat':1028 'turn':447 'two':1246,1250 'u':1006,1696 'u-shap':1005,1695 'undertrigg':110,643,1077,1119,1171,1195,1209,1491,1576,1634 'unrout':1648 'usag':207,1216,1593 'use':4,33,157,369,387,479,605,849,880,1380,1400 'useless':634 'user':7,37,115,272,356,393,405,444,527,598,658,682,697,700,715,721,1069,1141,1157,1174,1459,1482,1544,1615 'valu':1084 'valuabl':571 'varianc':1716 'via':372,457,1654 'violat':206,869 'vs':589,1677 'want':8,696 'week':81 'weight':1477,1551 'well':1669 'well-written':1668 'widen':1222 'within':495,1019 'without':554,1399 'word':98,353,637,897,901,929,932,1306,1320,1368,1396 'work':647,1469 'workflow':281,345,352,403,432,485,513,535,621,718,736,857,866,1248,1429,1485 'worth':652 'wrap':963 'write':188,1115,1658 'written':1670 'wrong':638,690 'x':150,943,1682 'x/n':804 'y':805 'yaml':337,956,968 'zhou':1046","prices":[{"id":"769de731-dd12-4e85-968a-23d57a713b32","listingId":"88d77435-25ec-4bd4-ae09-c9274d63c10d","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"hqhq1025","category":"skill-optimizer","install_from":"skills.sh"},"createdAt":"2026-04-18T22:16:10.683Z"}],"sources":[{"listingId":"88d77435-25ec-4bd4-ae09-c9274d63c10d","source":"github","sourceId":"hqhq1025/skill-optimizer/skill-optimizer","sourceUrl":"https://github.com/hqhq1025/skill-optimizer/tree/main/skills/skill-optimizer","isPrimary":false,"firstSeenAt":"2026-04-18T22:16:10.683Z","lastSeenAt":"2026-05-02T06:55:48.668Z"}],"details":{"listingId":"88d77435-25ec-4bd4-ae09-c9274d63c10d","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"hqhq1025","slug":"skill-optimizer","github":{"repo":"hqhq1025/skill-optimizer","stars":67,"topics":["agent-skills","agentskills","claude-code","claude-skill","codex","llm-agents","prompt-engineering","skill-optimizer"],"license":"mit","html_url":"https://github.com/hqhq1025/skill-optimizer","pushed_at":"2026-03-30T09:58:35Z","description":"Diagnose and optimize your Agent Skills (SKILL.md) with real session data + research-backed static analysis. Works with Claude Code, Codex, and any Agent Skills-compatible agent.","skill_md_sha":"974d4a7e2da925bdc952e563d2bf78aa129e6bec","skill_md_path":"skills/skill-optimizer/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/hqhq1025/skill-optimizer/tree/main/skills/skill-optimizer"},"layout":"multi","source":"github","category":"skill-optimizer","frontmatter":{"name":"skill-optimizer","description":"Use when the user wants to analyze, audit, or improve their Agent Skills (SKILL.md files). Triggers on /optimize-skill, /skill-audit, 'optimize skills', 'analyze skills', 'check my skills', 'skill quality'. Also use proactively when the user mentions skills aren't triggering, skills feel broken, or asks why a skill didn't fire."},"skills_sh_url":"https://skills.sh/hqhq1025/skill-optimizer/skill-optimizer"},"updatedAt":"2026-05-02T06:55:48.668Z"}}