{"id":"b14947e4-65af-4a5d-b14f-626c16af018e","shortId":"jMujez","kind":"skill","title":"agent-orchestration-improve-agent","tagline":"Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.","description":"# Agent Performance Optimization Workflow\n\nSystematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.\n\n[Extended thinking: Agent optimization requires a data-driven approach combining performance metrics, user feedback analysis, and advanced prompt engineering techniques. Success depends on systematic evaluation, targeted improvements, and rigorous testing with rollback capabilities for production safety.]\n\n## Use this skill when\n\n- Improving an existing agent's performance or reliability\n- Analyzing failure modes, prompt quality, or tool usage\n- Running structured A/B tests or evaluation suites\n- Designing iterative optimization workflows for agents\n\n## Do not use this skill when\n\n- You are building a brand-new agent from scratch\n- There are no metrics, feedback, or test cases available\n- The task is unrelated to agent performance or prompt quality\n\n## Instructions\n\n1. Establish baseline metrics and collect representative examples.\n2. Identify failure modes and prioritize high-impact fixes.\n3. Apply prompt and workflow improvements with measurable goals.\n4. Validate with tests and roll out changes in controlled stages.\n\n## Safety\n\n- Avoid deploying prompt changes without regression testing.\n- Roll back quickly if quality or safety metrics regress.\n\n## Phase 1: Performance Analysis and Baseline Metrics\n\nComprehensive analysis of agent performance using context-manager for historical data collection.\n\n### 1.1 Gather Performance Data\n\n```\nUse: context-manager\nCommand: analyze-agent-performance $ARGUMENTS --days 30\n```\n\nCollect metrics including:\n\n- Task completion rate (successful vs failed tasks)\n- Response accuracy and factual correctness\n- Tool usage efficiency (correct tools, call frequency)\n- Average response time and token consumption\n- User satisfaction indicators (corrections, retries)\n- Hallucination incidents and error patterns\n\n### 1.2 User Feedback Pattern Analysis\n\nIdentify recurring patterns in user interactions:\n\n- **Correction patterns**: Where users consistently modify outputs\n- **Clarification requests**: Common areas of ambiguity\n- **Task abandonment**: Points where users give up\n- **Follow-up questions**: Indicators of incomplete responses\n- **Positive feedback**: Successful patterns to preserve\n\n### 1.3 Failure Mode Classification\n\nCategorize failures by root cause:\n\n- **Instruction misunderstanding**: Role or task confusion\n- **Output format errors**: Structure or formatting issues\n- **Context loss**: Long conversation degradation\n- **Tool misuse**: Incorrect or inefficient tool selection\n- **Constraint violations**: Safety or business rule breaches\n- **Edge case handling**: Unusual input scenarios\n\n### 1.4 Baseline Performance Report\n\nGenerate quantitative baseline metrics:\n\n```\nPerformance Baseline:\n- Task Success Rate: [X%]\n- Average Corrections per Task: [Y]\n- Tool Call Efficiency: [Z%]\n- User Satisfaction Score: [1-10]\n- Average Response Latency: [Xms]\n- Token Efficiency Ratio: [X:Y]\n```\n\n## Phase 2: Prompt Engineering Improvements\n\nApply advanced prompt optimization techniques using prompt-engineer agent.\n\n### 2.1 Chain-of-Thought Enhancement\n\nImplement structured reasoning patterns:\n\n```\nUse: prompt-engineer\nTechnique: chain-of-thought-optimization\n```\n\n- Add explicit reasoning steps: \"Let's approach this step-by-step...\"\n- Include self-verification checkpoints: \"Before proceeding, verify that...\"\n- Implement recursive decomposition for complex tasks\n- Add reasoning trace visibility for debugging\n\n### 2.2 Few-Shot Example Optimization\n\nCurate high-quality examples from successful interactions:\n\n- **Select diverse examples** covering common use cases\n- **Include edge cases** that previously failed\n- **Show both positive and negative examples** with explanations\n- **Order examples** from simple to complex\n- **Annotate examples** with key decision points\n\nExample structure:\n\n```\nGood Example:\nInput: [User request]\nReasoning: [Step-by-step thought process]\nOutput: [Successful response]\nWhy this works: [Key success factors]\n\nBad Example:\nInput: [Similar request]\nOutput: [Failed response]\nWhy this fails: [Specific issues]\nCorrect approach: [Fixed version]\n```\n\n### 2.3 Role Definition Refinement\n\nStrengthen agent identity and capabilities:\n\n- **Core purpose**: Clear, single-sentence mission\n- **Expertise domains**: Specific knowledge areas\n- **Behavioral traits**: Personality and interaction style\n- **Tool proficiency**: Available tools and when to use them\n- **Constraints**: What the agent should NOT do\n- **Success criteria**: How to measure task completion\n\n### 2.4 Constitutional AI Integration\n\nImplement self-correction mechanisms:\n\n```\nConstitutional Principles:\n1. Verify factual accuracy before responding\n2. Self-check for potential biases or harmful content\n3. Validate output format matches requirements\n4. Ensure response completeness\n5. Maintain consistency with previous responses\n```\n\nAdd critique-and-revise loops:\n\n- Initial response generation\n- Self-critique against principles\n- Automatic revision if issues detected\n- Final validation before output\n\n### 2.5 Output Format Tuning\n\nOptimize response structure:\n\n- **Structured templates** for common tasks\n- **Dynamic formatting** based on complexity\n- **Progressive disclosure** for detailed information\n- **Markdown optimization** for readability\n- **Code block formatting** with syntax highlighting\n- **Table and list generation** for data presentation\n\n## Phase 3: Testing and Validation\n\nComprehensive testing framework with A/B comparison.\n\n### 3.1 Test Suite Development\n\nCreate representative test scenarios:\n\n```\nTest Categories:\n1. Golden path scenarios (common successful cases)\n2. Previously failed tasks (regression testing)\n3. Edge cases and corner scenarios\n4. Stress tests (complex, multi-step tasks)\n5. Adversarial inputs (potential breaking points)\n6. Cross-domain tasks (combining capabilities)\n```\n\n### 3.2 A/B Testing Framework\n\nCompare original vs improved agent:\n\n```\nUse: parallel-test-runner\nConfig:\n  - Agent A: Original version\n  - Agent B: Improved version\n  - Test set: 100 representative tasks\n  - Metrics: Success rate, speed, token usage\n  - Evaluation: Blind human review + automated scoring\n```\n\nStatistical significance testing:\n\n- Minimum sample size: 100 tasks per variant\n- Confidence level: 95% (p < 0.05)\n- Effect size calculation (Cohen's d)\n- Power analysis for future tests\n\n### 3.3 Evaluation Metrics\n\nComprehensive scoring framework:\n\n**Task-Level Metrics:**\n\n- Completion rate (binary success/failure)\n- Correctness score (0-100% accuracy)\n- Efficiency score (steps taken vs optimal)\n- Tool usage appropriateness\n- Response relevance and completeness\n\n**Quality Metrics:**\n\n- Hallucination rate (factual errors per response)\n- Consistency score (alignment with previous responses)\n- Format compliance (matches specified structure)\n- Safety score (constraint adherence)\n- User satisfaction prediction\n\n**Performance Metrics:**\n\n- Response latency (time to first token)\n- Total generation time\n- Token consumption (input + output)\n- Cost per task (API usage fees)\n- Memory/context efficiency\n\n### 3.4 Human Evaluation Protocol\n\nStructured human review process:\n\n- Blind evaluation (evaluators don't know version)\n- Standardized rubric with clear criteria\n- Multiple evaluators per sample (inter-rater reliability)\n- Qualitative feedback collection\n- Preference ranking (A vs B comparison)\n\n## Phase 4: Version Control and Deployment\n\nSafe rollout with monitoring and rollback capabilities.\n\n### 4.1 Version Management\n\nSystematic versioning strategy:\n\n```\nVersion Format: agent-name-v[MAJOR].[MINOR].[PATCH]\nExample: customer-support-v2.3.1\n\nMAJOR: Significant capability changes\nMINOR: Prompt improvements, new examples\nPATCH: Bug fixes, minor adjustments\n```\n\nMaintain version history:\n\n- Git-based prompt storage\n- Changelog with improvement details\n- Performance metrics per version\n- Rollback procedures documented\n\n### 4.2 Staged Rollout\n\nProgressive deployment strategy:\n\n1. **Alpha testing**: Internal team validation (5% traffic)\n2. **Beta testing**: Selected users (20% traffic)\n3. **Canary release**: Gradual increase (20% → 50% → 100%)\n4. **Full deployment**: After success criteria met\n5. **Monitoring period**: 7-day observation window\n\n### 4.3 Rollback Procedures\n\nQuick recovery mechanism:\n\n```\nRollback Triggers:\n- Success rate drops >10% from baseline\n- Critical errors increase >5%\n- User complaints spike\n- Cost per task increases >20%\n- Safety violations detected\n\nRollback Process:\n1. Detect issue via monitoring\n2. Alert team immediately\n3. Switch to previous stable version\n4. Analyze root cause\n5. Fix and re-test before retry\n```\n\n### 4.4 Continuous Monitoring\n\nReal-time performance tracking:\n\n- Dashboard with key metrics\n- Anomaly detection alerts\n- User feedback collection\n- Automated regression testing\n- Weekly performance reports\n\n## Success Criteria\n\nAgent improvement is successful when:\n\n- Task success rate improves by ≥15%\n- User corrections decrease by ≥25%\n- No increase in safety violations\n- Response time remains within 10% of baseline\n- Cost per task doesn't increase >5%\n- Positive user feedback increases\n\n## Post-Deployment Review\n\nAfter 30 days of production use:\n\n1. Analyze accumulated performance data\n2. Compare against baseline and targets\n3. Identify new improvement opportunities\n4. Document lessons learned\n5. Plan next optimization cycle\n\n## Continuous Improvement Cycle\n\nEstablish regular improvement cadence:\n\n- **Weekly**: Monitor metrics and collect feedback\n- **Monthly**: Analyze patterns and plan improvements\n- **Quarterly**: Major version updates with new capabilities\n- **Annually**: Strategic review and architecture updates\n\nRemember: Agent optimization is an iterative process. Each cycle builds upon previous learnings, gradually improving performance while maintaining stability and safety.\n\n## Limitations\n- Use this skill only when the task clearly matches the scope described above.\n- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.\n- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.","tags":["agent","orchestration","improve","antigravity","awesome","skills","sickn33","agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding"],"capabilities":["skill","source-sickn33","skill-agent-orchestration-improve-agent","topic-agent-skills","topic-agentic-skills","topic-ai-agent-skills","topic-ai-agents","topic-ai-coding","topic-ai-workflows","topic-antigravity","topic-antigravity-skills","topic-claude-code","topic-claude-code-skills","topic-codex-cli","topic-codex-skills"],"categories":["antigravity-awesome-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/sickn33/antigravity-awesome-skills/agent-orchestration-improve-agent","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add sickn33/antigravity-awesome-skills","source_repo":"https://github.com/sickn33/antigravity-awesome-skills","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 34997 github stars · SKILL.md body (10,258 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":"2026-04-25T06:50:23.049Z","embedding":null,"createdAt":"2026-04-18T21:30:29.231Z","updatedAt":"2026-04-25T06:50:23.049Z","lastSeenAt":"2026-04-25T06:50:23.049Z","tsv":"'-10':390 '-100':854 '0':853 '0.05':825 '1':142,198,389,616,731,1028,1096,1198 '1.1':217 '1.2':271 '1.3':316 '1.4':363 '10':1076,1174 '100':796,817,1050 '15':1159 '2':150,401,622,738,1036,1101,1203 '2.1':415 '2.2':468 '2.3':555 '2.4':605 '2.5':671 '20':1041,1048,1090 '25':1164 '3':160,632,711,744,1043,1105,1209 '3.1':721,988 '3.2':771 '3.3':837 '3.4':918 '30':232,1193 '4':169,638,750,956,1051,1111,1214 '4.1':968 '4.2':1022 '4.3':1065 '4.4':1123 '5':642,758,1034,1058,1082,1115,1183,1218 '50':1049 '6':764 '7':1061 '95':823 'a/b':95,719,772 'abandon':296 'accumul':1200 'accuraci':244,619,855 'add':435,462,648 'adher':891 'adjust':1002 'advanc':53,406 'adversari':759 'agent':2,5,10,19,27,38,80,105,119,136,207,228,414,560,594,779,786,790,977,1149,1256 'agent-name-v':976 'agent-orchestration-improve-ag':1 'ai':607 'alert':1102,1137 'align':879 'alpha':1029 'ambigu':294 'analysi':13,30,51,200,205,275,833 'analyz':85,227,1112,1199,1237 'analyze-agent-perform':226 'annot':509 'annual':1249 'anomali':1135 'api':913 'appli':161,405 'approach':45,441,552 'appropri':864 'architectur':1253 'area':292,575 'argument':230 'ask':1309 'autom':809,1141 'automat':662 'avail':130,584 'averag':255,377,391 'avoid':181 'b':791,953 'back':189 'bad':538 'base':685,1008 'baselin':144,202,364,369,372,1078,1176,1206 'behavior':576 'beta':1037 'bias':628 'binari':849 'blind':806,926 'block':698 'boundari':1317 'brand':117 'brand-new':116 'breach':356 'break':762 'bug':999 'build':114,1264 'busi':354 'cadenc':1229 'calcul':828 'call':253,383 'canari':1044 'capabl':69,563,770,967,991,1248 'case':129,358,488,491,737,746 'categor':320 'categori':730 'caus':324,1114 'chain':417,431 'chain-of-thought':416 'chain-of-thought-optim':430 'chang':176,184,992 'changelog':1011 'check':625 'checkpoint':451 'clarif':289,1311 'classif':319 'clear':566,936,1284 'code':697 'cohen':829 'collect':147,216,233,948,1140,1234 'combin':46,769 'command':225 'common':291,486,681,735 'compar':775,1204 'comparison':720,954 'complaint':1084 'complet':237,604,641,847,868 'complex':460,508,687,753 'complianc':884 'comprehens':204,715,840 'confid':821 'config':785 'confus':330 'consist':286,644,877 'constitut':606,614 'constraint':350,591,890 'consumpt':260,907 'content':631 'context':211,223,338 'context-manag':210,222 'continu':17,34,1124,1223 'control':178,958 'convers':341 'core':564 'corner':748 'correct':247,251,264,282,378,551,612,851,1161 'cost':910,1086,1177 'cover':485 'creat':725 'criteria':599,937,1056,1148,1320 'critic':1079 'critiqu':650,659 'critique-and-revis':649 'cross':766 'cross-domain':765 'curat':474 'custom':985 'customer-support-v2':984 'cycl':1222,1225,1263 'd':831 'dashboard':1131 'data':43,215,220,708,1202 'data-driven':42 'day':231,1062,1194 'debug':467 'decis':513 'decomposit':458 'decreas':1162 'definit':557 'degrad':342 'depend':58 'deploy':182,960,1026,1053,1190 'describ':1288 'design':100 'detail':691,1014 'detect':666,1093,1097,1136 'develop':724 'disclosur':689 'divers':483 'document':1021,1215 'doesn':1180 'domain':572,767 'driven':44 'drop':1075 'dynam':683 'edg':357,490,745 'effect':826 'effici':250,384,396,856,917 'engin':15,32,55,403,413,428 'enhanc':420 'ensur':639 'environ':1300 'environment-specif':1299 'error':269,333,874,1080 'establish':143,1226 'evalu':61,98,805,838,920,927,928,939 'exampl':149,472,478,484,500,504,510,515,518,539,983,997 'exist':9,26,79 'expert':1305 'expertis':571 'explan':502 'explicit':436 'extend':36 'factor':537 'factual':246,618,873 'fail':241,494,544,548,740 'failur':86,152,317,321 'fee':915 'feedback':50,126,273,311,947,1139,1186,1235 'few-shot':469 'final':667 'first':901 'fix':159,553,1000,1116 'follow':303 'follow-up':302 'format':332,336,635,673,684,699,883,975 'framework':717,774,842 'frequenc':254 'full':1052 'futur':835 'gather':218 'generat':367,656,706,904 'git':1007 'git-bas':1006 'give':300 'goal':168 'golden':732 'good':517 'gradual':1046,1268 'hallucin':266,871 'handl':359 'harm':630 'high':157,476 'high-impact':156 'high-qual':475 'highlight':702 'histor':214 'histori':1005 'human':807,919,923 'ident':561 'identifi':151,276,1210 'immedi':1104 'impact':158 'implement':421,456,609 'improv':4,7,24,63,77,165,404,778,792,995,1013,1150,1157,1212,1224,1228,1241,1269 'incid':267 'includ':235,447,489 'incomplet':308 'incorrect':345 'increas':1047,1081,1089,1166,1182,1187 'indic':263,306 'ineffici':347 'inform':692 'initi':654 'input':361,519,540,760,908,1314 'instruct':141,325 'integr':608 'inter':943 'inter-rat':942 'interact':281,481,580 'intern':1031 'issu':337,550,665,1098 'iter':18,35,101,1260 'key':512,535,1133 'know':931 'knowledg':574 'latenc':393,898 'learn':1217,1267 'lesson':1216 'let':439 'level':822,845 'limit':1276 'list':705 'long':340 'loop':653 'loss':339 'maintain':643,1003,1272 'major':980,989,1243 'manag':212,224,970 'markdown':693 'match':636,885,1285 'measur':167,602 'mechan':613,1070 'memory/context':916 'met':1057 'metric':48,125,145,195,203,234,370,799,839,846,870,896,1016,1134,1232 'minimum':814 'minor':981,993,1001 'miss':1322 'mission':570 'misunderstand':326 'misus':344 'mode':87,153,318 'modifi':287 'monitor':964,1059,1100,1125,1231 'month':1236 'multi':755 'multi-step':754 'multipl':938 'name':978 'negat':499 'new':118,996,1211,1247 'next':1220 'observ':1063 'opportun':1213 'optim':21,39,102,408,434,473,675,694,861,1221,1257 'orchestr':3 'order':503 'origin':776,788 'output':288,331,529,543,634,670,672,909,1294 'p':824 'parallel':782 'parallel-test-runn':781 'patch':982,998 'path':733 'pattern':270,274,278,283,313,424,1238 'per':379,819,875,911,940,1017,1087,1178 'perform':12,20,29,47,82,137,199,208,219,229,365,371,895,1015,1129,1145,1201,1270 'period':1060 'permiss':1315 'person':578 'phase':197,400,710,955 'plan':1219,1240 'point':297,514,763 'posit':310,497,1184 'post':1189 'post-deploy':1188 'potenti':627,761 'power':832 'predict':894 'prefer':949 'present':709 'preserv':315 'previous':493,646,739,881,1108,1266 'principl':615,661 'priorit':155 'procedur':1020,1067 'proceed':453 'process':528,925,1095,1261 'product':71,1196 'profici':583 'progress':688,1025 'prompt':14,31,54,88,139,162,183,402,407,412,427,994,1009 'prompt-engin':411,426 'protocol':921 'purpos':565 'qualit':946 'qualiti':89,140,192,477,869 'quantit':368 'quarter':1242 'question':305 'quick':190,1068 'rank':950 'rate':238,375,801,848,872,1074,1156 'rater':944 'ratio':397 're':1119 're-test':1118 'readabl':696 'real':1127 'real-tim':1126 'reason':423,437,463,522 'recoveri':1069 'recur':277 'recurs':457 'refin':558 'regress':186,196,742,1142 'regular':1227 'releas':1045 'relev':866 'reliabl':84,945 'remain':1172 'rememb':1255 'report':366,1146 'repres':148,726,797 'request':290,521,542 'requir':40,637,1313 'respond':621 'respons':243,256,309,392,531,545,640,647,655,676,865,876,882,897,1170 'retri':265,1122 'review':808,924,1191,1251,1306 'revis':652,663 'rigor':65 'role':327,556 'roll':174,188 'rollback':68,966,1019,1066,1071,1094 'rollout':962,1024 'root':323,1113 'rubric':934 'rule':355 'run':93 'runner':784 'safe':961 'safeti':72,180,194,352,888,1091,1168,1275,1316 'sampl':815,941 'satisfact':262,387,893 'scenario':362,728,734,749 'scope':1287 'score':388,810,841,852,857,878,889 'scratch':121 'select':349,482,1039 'self':449,611,624,658 'self-check':623 'self-correct':610 'self-critiqu':657 'self-verif':448 'sentenc':569 'set':795 'shot':471 'show':495 'signific':812,990 'similar':541 'simpl':506 'singl':568 'single-sent':567 'size':816,827 'skill':75,110,1279 'skill-agent-orchestration-improve-agent' 'source-sickn33' 'specif':549,573,1301 'specifi':886 'speed':802 'spike':1085 'stabil':1273 'stabl':1109 'stage':179,1023 'standard':933 'statist':811 'step':438,444,446,524,526,756,858 'step-by-step':443,523 'stop':1307 'storag':1010 'strateg':1250 'strategi':973,1027 'strengthen':559 'stress':751 'structur':94,334,422,516,677,678,887,922 'style':581 'substitut':1297 'success':57,239,312,374,480,530,536,598,736,800,1055,1073,1147,1152,1155,1319 'success/failure':850 'suit':99,723 'support':986 'switch':1106 'syntax':701 'systemat':6,23,60,971 'tabl':703 'taken':859 'target':62,1208 'task':132,236,242,295,329,373,380,461,603,682,741,757,768,798,818,844,912,1088,1154,1179,1283 'task-level':843 'team':1032,1103 'techniqu':56,409,429 'templat':679 'test':66,96,128,172,187,712,716,722,727,729,743,752,773,783,794,813,836,1030,1038,1120,1143,1303 'think':37 'thought':419,433,527 'time':257,899,905,1128,1171 'token':259,395,803,902,906 'tool':91,248,252,343,348,382,582,585,862 'topic-agent-skills' 'topic-agentic-skills' 'topic-ai-agent-skills' 'topic-ai-agents' 'topic-ai-coding' 'topic-ai-workflows' 'topic-antigravity' 'topic-antigravity-skills' 'topic-claude-code' 'topic-claude-code-skills' 'topic-codex-cli' 'topic-codex-skills' 'total':903 'trace':464 'track':1130 'traffic':1035,1042 'trait':577 'treat':1292 'trigger':1072 'tune':674 'unrel':134 'unusu':360 'updat':1245,1254 'upon':1265 'usag':92,249,804,863,914 'use':73,108,209,221,410,425,487,589,780,1197,1277 'user':49,261,272,280,285,299,386,520,892,1040,1083,1138,1160,1185 'v':979 'v2':987 'valid':170,633,668,714,1033,1302 'variant':820 'verif':450 'verifi':454,617 'version':554,789,793,932,957,969,972,974,1004,1018,1110,1244 'via':1099 'violat':351,1092,1169 'visibl':465 'vs':240,777,860,952 'week':1144,1230 'window':1064 'within':1173 'without':185 'work':534 'workflow':22,103,164 'x':376,398 'xms':394 'y':381,399 'z':385","prices":[{"id":"d6682af1-0c02-4bbf-9947-0e85e58f0004","listingId":"b14947e4-65af-4a5d-b14f-626c16af018e","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"sickn33","category":"antigravity-awesome-skills","install_from":"skills.sh"},"createdAt":"2026-04-18T21:30:29.231Z"}],"sources":[{"listingId":"b14947e4-65af-4a5d-b14f-626c16af018e","source":"github","sourceId":"sickn33/antigravity-awesome-skills/agent-orchestration-improve-agent","sourceUrl":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/agent-orchestration-improve-agent","isPrimary":false,"firstSeenAt":"2026-04-18T21:30:29.231Z","lastSeenAt":"2026-04-25T06:50:23.049Z"}],"details":{"listingId":"b14947e4-65af-4a5d-b14f-626c16af018e","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"sickn33","slug":"agent-orchestration-improve-agent","github":{"repo":"sickn33/antigravity-awesome-skills","stars":34997,"topics":["agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding","ai-workflows","antigravity","antigravity-skills","claude-code","claude-code-skills","codex-cli","codex-skills","cursor","cursor-skills","developer-tools","gemini-cli","gemini-skills","kiro","mcp","skill-library"],"license":"mit","html_url":"https://github.com/sickn33/antigravity-awesome-skills","pushed_at":"2026-04-25T06:33:17Z","description":"Installable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.","skill_md_sha":"804a7b7b8e0f417854805904f93c7d10aabce5ce","skill_md_path":"skills/agent-orchestration-improve-agent/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/agent-orchestration-improve-agent"},"layout":"multi","source":"github","category":"antigravity-awesome-skills","frontmatter":{"name":"agent-orchestration-improve-agent","description":"Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration."},"skills_sh_url":"https://skills.sh/sickn33/antigravity-awesome-skills/agent-orchestration-improve-agent"},"updatedAt":"2026-04-25T06:50:23.049Z"}}