{"id":"1a1d8899-0773-4589-9612-5a8420729312","shortId":"jTRmNk","kind":"skill","title":"ai-agents-architect","tagline":"Expert in designing and building autonomous AI agents. Masters tool","description":"# AI Agents Architect\n\nExpert in designing and building autonomous AI agents. Masters tool use,\nmemory systems, planning strategies, and multi-agent orchestration.\n\n**Role**: AI Agent Systems Architect\n\nI build AI systems that can act autonomously while remaining controllable.\nI understand that agents fail in unexpected ways - I design for graceful\ndegradation and clear failure modes. I balance autonomy with oversight,\nknowing when an agent should ask for help vs proceed independently.\n\n### Expertise\n\n- Agent loop design (ReAct, Plan-and-Execute, etc.)\n- Tool definition and execution\n- Memory architectures (short-term, long-term, episodic)\n- Planning strategies and task decomposition\n- Multi-agent communication patterns\n- Agent evaluation and observability\n- Error handling and recovery\n- Safety and guardrails\n\n### Principles\n\n- Agents should fail loudly, not silently\n- Every tool needs clear documentation and examples\n- Memory is for context, not crutch\n- Planning reduces but doesn't eliminate errors\n- Multi-agent adds complexity - justify the overhead\n\n## Capabilities\n\n- Agent architecture design\n- Tool and function calling\n- Agent memory systems\n- Planning and reasoning strategies\n- Multi-agent orchestration\n- Agent evaluation and debugging\n\n## Prerequisites\n\n- Required skills: LLM API usage, Understanding of function calling, Basic prompt engineering\n\n## Patterns\n\n### ReAct Loop\n\nReason-Act-Observe cycle for step-by-step execution\n\n**When to use**: Simple tool use with clear action-observation flow\n\n- Thought: reason about what to do next\n- Action: select and invoke a tool\n- Observation: process tool result\n- Repeat until task complete or stuck\n- Include max iteration limits\n\n### Plan-and-Execute\n\nPlan first, then execute steps\n\n**When to use**: Complex tasks requiring multi-step planning\n\n- Planning phase: decompose task into steps\n- Execution phase: execute each step\n- Replanning: adjust plan based on results\n- Separate planner and executor models possible\n\n### Tool Registry\n\nDynamic tool discovery and management\n\n**When to use**: Many tools or tools that change at runtime\n\n- Register tools with schema and examples\n- Tool selector picks relevant tools for task\n- Lazy loading for expensive tools\n- Usage tracking for optimization\n\n### Hierarchical Memory\n\nMulti-level memory for different purposes\n\n**When to use**: Long-running agents needing context\n\n- Working memory: current task context\n- Episodic memory: past interactions/results\n- Semantic memory: learned facts and patterns\n- Use RAG for retrieval from long-term memory\n\n### Supervisor Pattern\n\nSupervisor agent orchestrates specialist agents\n\n**When to use**: Complex tasks requiring multiple skills\n\n- Supervisor decomposes and delegates\n- Specialists have focused capabilities\n- Results aggregated by supervisor\n- Error handling at supervisor level\n\n### Checkpoint Recovery\n\nSave state for resumption after failures\n\n**When to use**: Long-running tasks that may fail\n\n- Checkpoint after each successful step\n- Store task state, memory, and progress\n- Resume from last checkpoint on failure\n- Clean up checkpoints on completion\n\n## Sharp Edges\n\n### Agent loops without iteration limits\n\nSeverity: CRITICAL\n\nSituation: Agent runs until 'done' without max iterations\n\nSymptoms:\n- Agent runs forever\n- Unexplained high API costs\n- Application hangs\n\nWhy this breaks:\nAgents can get stuck in loops, repeating the same actions, or spiral\ninto endless tool calls. Without limits, this drains API credits,\nhangs the application, and frustrates users.\n\nRecommended fix:\n\nAlways set limits:\n- max_iterations on agent loops\n- max_tokens per turn\n- timeout on agent runs\n- cost caps for API usage\n- Circuit breakers for tool failures\n\n### Vague or incomplete tool descriptions\n\nSeverity: HIGH\n\nSituation: Tool descriptions don't explain when/how to use\n\nSymptoms:\n- Agent picks wrong tools\n- Parameter errors\n- Agent says it can't do things it can\n\nWhy this breaks:\nAgents choose tools based on descriptions. Vague descriptions lead to\nwrong tool selection, misused parameters, and errors. The agent\nliterally can't know what it doesn't see in the description.\n\nRecommended fix:\n\nWrite complete tool specs:\n- Clear one-sentence purpose\n- When to use (and when not to)\n- Parameter descriptions with types\n- Example inputs and outputs\n- Error cases to expect\n\n### Tool errors not surfaced to agent\n\nSeverity: HIGH\n\nSituation: Catching tool exceptions silently\n\nSymptoms:\n- Agent continues with wrong data\n- Final answers are wrong\n- Hard to debug failures\n\nWhy this breaks:\nWhen tool errors are swallowed, the agent continues with bad or missing\ndata, compounding errors. The agent can't recover from what it can't\nsee. Silent failures become loud failures later.\n\nRecommended fix:\n\nExplicit error handling:\n- Return error messages to agent\n- Include error type and recovery hints\n- Let agent retry or choose alternative\n- Log errors for debugging\n\n### Storing everything in agent memory\n\nSeverity: MEDIUM\n\nSituation: Appending all observations to memory without filtering\n\nSymptoms:\n- Context window exceeded\n- Agent references outdated info\n- High token costs\n\nWhy this breaks:\nMemory fills with irrelevant details, old information, and noise.\nThis bloats context, increases costs, and can cause the model to\nlose focus on what matters.\n\nRecommended fix:\n\nSelective memory:\n- Summarize rather than store verbatim\n- Filter by relevance before storing\n- Use RAG for long-term memory\n- Clear working memory between tasks\n\n### Agent has too many tools\n\nSeverity: MEDIUM\n\nSituation: Giving agent 20+ tools for flexibility\n\nSymptoms:\n- Wrong tool selection\n- Agent overwhelmed by options\n- Slow responses\n\nWhy this breaks:\nMore tools means more confusion. The agent must read and consider all\ntool descriptions, increasing latency and error rate. Long tool lists\nget cut off or poorly understood.\n\nRecommended fix:\n\nCurate tools per task:\n- 5-10 tools maximum per agent\n- Use tool selection layer for large tool sets\n- Specialized agents with focused tools\n- Dynamic tool loading based on task\n\n### Using multiple agents when one would work\n\nSeverity: MEDIUM\n\nSituation: Starting with multi-agent architecture for simple tasks\n\nSymptoms:\n- Agents duplicating work\n- Communication overhead\n- Hard to debug failures\n\nWhy this breaks:\nMulti-agent adds coordination overhead, communication failures,\ndebugging complexity, and cost. Each agent handoff is a potential\nfailure point. Start simple, add agents only when proven necessary.\n\nRecommended fix:\n\nJustify multi-agent:\n- Can one agent with good tools solve this?\n- Is the coordination overhead worth it?\n- Are the agents truly independent?\n- Start with single agent, measure limits\n\n### Agent internals not logged or traceable\n\nSeverity: MEDIUM\n\nSituation: Running agents without logging thoughts/actions\n\nSymptoms:\n- Can't explain agent failures\n- No visibility into agent reasoning\n- Debugging takes hours\n\nWhy this breaks:\nWhen agents fail, you need to see what they were thinking, which\ntools they tried, and where they went wrong. Without observability,\ndebugging is guesswork.\n\nRecommended fix:\n\nImplement tracing:\n- Log each thought/action/observation\n- Track tool calls with inputs/outputs\n- Trace token usage and latency\n- Use structured logging for analysis\n\n### Fragile parsing of agent outputs\n\nSeverity: MEDIUM\n\nSituation: Regex or exact string matching on LLM output\n\nSymptoms:\n- Parse errors in agent loop\n- Works sometimes, fails sometimes\n- Small prompt changes break parsing\n\nWhy this breaks:\nLLMs don't produce perfectly consistent output. Minor format variations\nbreak brittle parsers. This causes agent crashes or incorrect behavior\nfrom parsing errors.\n\nRecommended fix:\n\nRobust output handling:\n- Use structured output (JSON mode, function calling)\n- Fuzzy matching for actions\n- Retry with format instructions on parse failure\n- Handle multiple output formats\n\n## Related Skills\n\nWorks well with: `rag-engineer`, `prompt-engineer`, `backend`, `mcp-builder`\n\n## When to Use\n- User mentions or implies: build agent\n- User mentions or implies: AI agent\n- User mentions or implies: autonomous agent\n- User mentions or implies: tool use\n- User mentions or implies: function calling\n- User mentions or implies: multi-agent\n- User mentions or implies: agent memory\n- User mentions or implies: agent planning\n- User mentions or implies: langchain agent\n- User mentions or implies: crewai\n- User mentions or implies: autogen\n- User mentions or implies: claude agent sdk\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":["agents","architect","antigravity","awesome","skills","sickn33","agent-skills","agentic-skills","ai-agent-skills","ai-agents","ai-coding","ai-workflows"],"capabilities":["skill","source-sickn33","skill-ai-agents-architect","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/ai-agents-architect","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 · 37911 github stars · SKILL.md body (8,841 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":"2026-05-18T18:50:25.992Z","embedding":null,"createdAt":"2026-04-18T20:35:04.780Z","updatedAt":"2026-05-18T18:50:25.992Z","lastSeenAt":"2026-05-18T18:50:25.992Z","tsv":"'-10':863 '20':811 '5':862 'act':49,207 'action':225,235,490,1128 'action-observ':224 'add':161,922,941 'adjust':286 'agent':3,12,16,25,36,40,57,79,88,117,120,132,160,167,174,183,185,352,382,385,453,461,469,481,517,525,554,560,572,590,638,647,669,679,704,712,724,740,801,810,819,834,867,877,889,901,907,921,932,942,952,955,969,975,978,988,996,1001,1010,1059,1076,1105,1163,1169,1175,1194,1199,1205,1212,1228 'aggreg':403 'ai':2,11,15,24,39,45,1168 'ai-agents-architect':1 'altern':716 'alway':511 'analysi':1055 'answer':653 'api':193,474,501,530 'append':729 'applic':476,505 'architect':4,17,42 'architectur':102,168,902 'ask':81,1263 'autogen':1222 'autonom':10,23,50,1174 'autonomi':73 'backend':1151 'bad':672 'balanc':72 'base':288,575,884 'basic':199 'becom':691 'behavior':1109 'bloat':760 'boundari':1271 'break':480,571,662,749,827,918,1008,1085,1089,1100 'breaker':533 'brittl':1101 'build':9,22,44,1162 'builder':1154 'call':173,198,496,1043,1124,1187 'cap':528 'capabl':166,401 'case':630 'catch':642 'caus':766,1104 'chang':312,1084 'checkpoint':411,429,443,448 'choos':573,715 'circuit':532 'clarif':1265 'claud':1227 'clean':446 'clear':68,141,223,609,796,1238 'communic':118,910,925 'complet':248,450,606 'complex':162,267,389,928 'compound':676 'confus':832 'consid':838 'consist':1095 'context':148,354,359,737,761 'continu':648,670 'control':53 'coordin':923,963 'cost':475,527,746,763,930 'crash':1106 'credit':502 'crewai':1217 'criteria':1274 'critic':459 'crutch':150 'curat':858 'current':357 'cut':851 'cycl':209 'data':651,675 'debug':188,658,720,914,927,1003,1031 'decompos':276,395 'decomposit':114 'definit':98 'degrad':66 'deleg':397 'describ':1242 'descript':541,546,577,579,602,622,841 'design':7,20,63,90,169 'detail':754 'differ':344 'discoveri':301 'document':142 'doesn':154,597 'done':464 'drain':500 'duplic':908 'dynam':299,881 'edg':452 'elimin':156 'endless':494 'engin':201,1147,1150 'environ':1254 'environment-specif':1253 'episod':109,360 'error':124,157,406,559,588,629,634,665,677,698,701,706,718,845,1074,1112 'etc':96 'evalu':121,186 'everi':138 'everyth':722 'exact':1066 'exampl':144,320,625 'exceed':739 'except':644 'execut':95,100,215,258,262,280,282 'executor':294 'expect':632 'expens':331 'expert':5,18,1259 'expertis':87 'explain':549,995 'explicit':697 'fact':367 'fail':58,134,428,1011,1080 'failur':69,418,445,536,659,690,693,915,926,937,997,1135 'fill':751 'filter':735,784 'final':652 'first':260 'fix':510,604,696,776,857,948,1035,1114 'flexibl':814 'flow':227 'focus':400,771,879 'forev':471 'format':1098,1131,1139 'fragil':1056 'frustrat':507 'function':172,197,1123,1186 'fuzzi':1125 'get':483,850 'give':809 'good':957 'grace':65 'guardrail':130 'guesswork':1033 'handl':125,407,699,1117,1136 'handoff':933 'hang':477,503 'hard':656,912 'help':83 'hierarch':337 'high':473,543,640,744 'hint':710 'hour':1005 'implement':1036 'impli':1161,1167,1173,1179,1185,1191,1198,1204,1210,1216,1221,1226 'includ':251,705 'incomplet':539 'incorrect':1108 'increas':762,842 'independ':86,971 'info':743 'inform':756 'input':626,1268 'inputs/outputs':1045 'instruct':1132 'interactions/results':363 'intern':979 'invok':238 'irrelev':753 'iter':253,456,467,515 'json':1121 'justifi':163,949 'know':76,594 'langchain':1211 'larg':873 'last':442 'latenc':843,1050 'later':694 'layer':871 'lazi':328 'lead':580 'learn':366 'let':711 'level':341,410 'limit':254,457,498,513,977,1230 'list':849 'liter':591 'llm':192,1070 'llms':1090 'load':329,883 'log':717,981,990,1038,1053 'long':107,350,376,423,793,847 'long-run':349,422 'long-term':106,375,792 'loop':89,204,454,486,518,1077 'lose':770 'loud':135,692 'manag':303 'mani':307,804 'master':13,26 'match':1068,1126,1239 'matter':774 'max':252,466,514,519 'maximum':865 'may':427 'mcp':1153 'mcp-builder':1152 'mean':830 'measur':976 'medium':727,807,895,985,1062 'memori':29,101,145,175,338,342,356,361,365,378,437,725,733,750,778,795,798,1200 'mention':1159,1165,1171,1177,1183,1189,1196,1202,1208,1214,1219,1224 'messag':702 'minor':1097 'miss':674,1276 'misus':585 'mode':70,1122 'model':295,768 'multi':35,116,159,182,271,340,900,920,951,1193 'multi-ag':34,115,158,181,899,919,950,1192 'multi-level':339 'multi-step':270 'multipl':392,888,1137 'must':835 'necessari':946 'need':140,353,1013 'next':234 'nois':758 'observ':123,208,226,241,731,1030 'old':755 'one':611,891,954 'one-sent':610 'optim':336 'option':822 'orchestr':37,184,383 'outdat':742 'output':628,1060,1071,1096,1116,1120,1138,1248 'overhead':165,911,924,964 'oversight':75 'overwhelm':820 'paramet':558,586,621 'pars':1057,1073,1086,1111,1134 'parser':1102 'past':362 'pattern':119,202,369,380 'per':521,860,866 'perfect':1094 'permiss':1269 'phase':275,281 'pick':323,555 'plan':31,93,110,151,177,256,259,273,274,287,1206 'plan-and-execut':92,255 'planner':292 'point':938 'poor':854 'possibl':296 'potenti':936 'prerequisit':189 'principl':131 'proceed':85 'process':242 'produc':1093 'progress':439 'prompt':200,1083,1149 'prompt-engin':1148 'proven':945 'purpos':345,613 'rag':371,790,1146 'rag-engin':1145 'rate':846 'rather':780 'react':91,203 'read':836 'reason':179,206,229,1002 'reason-act-observ':205 'recommend':509,603,695,775,856,947,1034,1113 'recov':682 'recoveri':127,412,709 'reduc':152 'refer':741 'regex':1064 'regist':315 'registri':298 'relat':1140 'relev':324,786 'remain':52 'repeat':245,487 'replan':285 'requir':190,269,391,1267 'respons':824 'result':244,290,402 'resum':440 'resumpt':416 'retri':713,1129 'retriev':373 'return':700 'review':1260 'robust':1115 'role':38 'run':351,424,462,470,526,987 'runtim':314 'safeti':128,1270 'save':413 'say':561 'schema':318 'scope':1241 'sdk':1229 'see':599,688,1015 'select':236,584,777,818,870 'selector':322 'semant':364 'sentenc':612 'separ':291 'set':512,875 'sever':458,542,639,726,806,894,984,1061 'sharp':451 'short':104 'short-term':103 'silent':137,645,689 'simpl':219,904,940 'singl':974 'situat':460,544,641,728,808,896,986,1063 'skill':191,393,1141,1233 'skill-ai-agents-architect' 'slow':823 'small':1082 'solv':959 'sometim':1079,1081 'source-sickn33' 'spec':608 'special':876 'specialist':384,398 'specif':1255 'spiral':492 'start':897,939,972 'state':414,436 'step':212,214,263,272,279,284,433 'step-by-step':211 'stop':1261 'store':434,721,782,788 'strategi':32,111,180 'string':1067 'structur':1052,1119 'stuck':250,484 'substitut':1251 'success':432,1273 'summar':779 'supervisor':379,381,394,405,409 'surfac':636 'swallow':667 'symptom':468,553,646,736,815,906,992,1072 'system':30,41,46,176 'take':1004 'task':113,247,268,277,327,358,390,425,435,800,861,886,905,1237 'term':105,108,377,794 'test':1257 'thing':566 'think':1019 'thought':228 'thought/action/observation':1040 'thoughts/actions':991 'timeout':523 'token':520,745,1047 'tool':14,27,97,139,170,220,240,243,297,300,308,310,316,321,325,332,495,535,540,545,557,574,583,607,633,643,664,805,812,817,829,840,848,859,864,869,874,880,882,958,1021,1042,1180 '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' 'trace':1037,1046 'traceabl':983 'track':334,1041 'treat':1246 'tri':1023 'truli':970 'turn':522 'type':624,707 'understand':55,195 'understood':855 'unexpect':60 'unexplain':472 'usag':194,333,531,1048 'use':28,218,221,266,306,348,370,388,421,552,616,789,868,887,1051,1118,1157,1181,1231 'user':508,1158,1164,1170,1176,1182,1188,1195,1201,1207,1213,1218,1223 'vagu':537,578 'valid':1256 'variat':1099 'verbatim':783 'visibl':999 'vs':84 'way':61 'well':1143 'went':1027 'when/how':550 'window':738 'without':455,465,497,734,989,1029 'work':355,797,893,909,1078,1142 'worth':965 'would':892 'write':605 'wrong':556,582,650,655,816,1028","prices":[{"id":"360746a4-c26f-4eb5-9a95-5f4c2d0f9ad9","listingId":"1a1d8899-0773-4589-9612-5a8420729312","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":"2026-04-18T20:35:04.780Z"}],"sources":[{"listingId":"1a1d8899-0773-4589-9612-5a8420729312","source":"github","sourceId":"sickn33/antigravity-awesome-skills/ai-agents-architect","sourceUrl":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/ai-agents-architect","isPrimary":false,"firstSeenAt":"2026-04-18T21:30:38.384Z","lastSeenAt":"2026-05-18T18:50:25.992Z"},{"listingId":"1a1d8899-0773-4589-9612-5a8420729312","source":"skills_sh","sourceId":"sickn33/antigravity-awesome-skills/ai-agents-architect","sourceUrl":"https://skills.sh/sickn33/antigravity-awesome-skills/ai-agents-architect","isPrimary":true,"firstSeenAt":"2026-04-18T20:35:04.780Z","lastSeenAt":"2026-05-07T22:40:40.758Z"}],"details":{"listingId":"1a1d8899-0773-4589-9612-5a8420729312","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"sickn33","slug":"ai-agents-architect","github":{"repo":"sickn33/antigravity-awesome-skills","stars":37911,"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":"2026-05-18T08:24:49Z","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":"79d9c4b4e9b128a60fb820a6c215cd1be5167c77","skill_md_path":"skills/ai-agents-architect/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/ai-agents-architect"},"layout":"multi","source":"github","category":"antigravity-awesome-skills","frontmatter":{"name":"ai-agents-architect","description":"Expert in designing and building autonomous AI agents. Masters tool"},"skills_sh_url":"https://skills.sh/sickn33/antigravity-awesome-skills/ai-agents-architect"},"updatedAt":"2026-05-18T18:50:25.992Z"}}