{"id":"59887f75-b072-4750-b600-1544c01ae5ed","shortId":"jRRL5G","kind":"skill","title":"meta-prompting","tagline":">-","description":"# Meta-Prompting\n\nEnhanced reasoning via `/commands` or natural language. Commands combine left-to-right: `/verify /adversarial`. Auto-trigger when context warrants -- note which pattern applied. Output: apply the pattern inline, then mark the result (e.g., `VERIFIED ANSWER:`, `REVISED ANSWER:`, confidence tier).\n\n## Patterns\n\n**`/think`** | `/show` -- Show reasoning step-by-step: decision points, alternatives considered, why each accepted/rejected. With `/think doubt`: after each step, flag what could be wrong and why before proceeding.\n\n**`/adversarial`** | `/argue` -- After answering, steelman the opposing case. 3 strongest counterarguments ranked by severity. Identify blind spots and unstated assumptions.\n\n**`/constrain`** | `/strict` -- Tight constraints: 3 sentences max, cite sources, no hedging. Override inline: `/constrain 5 sentences`.\n\n**`/json`** | `/format` -- Respond in valid JSON code block, no surrounding prose unless asked. Default schema:\n```json\n{\"analysis\": \"string\", \"confidence_score\": 85, \"methodology\": \"string\", \"limitations\": [\"string\"]}\n```\nCustom keys: `/json {keys: summary, risks, recommendation}`\n\n**`/budget`** | `/deep` -- Extended thinking space (~500 words) showing dead ends and reasoning pivots, then clearly separated final answer.\n\n**`/compare`** | `/vs` -- Compare options as table. Default dimensions: speed, accuracy, cost, complexity, maintenance. Custom: `/compare [dim1, dim2]`.\n\n**`/confidence`** | `/conf` -- Rate each claim 0-100. Flag below 70 as SPECULATIVE. Group by tier: HIGH (85+), MEDIUM (70-84), LOW (<70). Include assumptions made and rate each 1-10 on confidence.\n\n**`/edge`** | `/break` -- 5+ inputs/scenarios that break the approach. Code: null/empty, concurrency, overflow, encoding, auth bypass. Strategies: market conditions, timing, dependencies.\n*Auto-triggers on: security, validation, parsing contexts.*\n\n**`/verify-think`** | `/check` -- Three phases: (1) **Answer** direct response, (2) **Challenge** 3 ways it could be wrong, (3) **Verify** investigate each, update if needed. Mark final as `VERIFIED ANSWER:` or `REVISED ANSWER:`. Distinct from the `/ia-verify` slash command, which runs the full pre-PR verification pipeline.\n*Auto-triggers on: architecture decisions, critical choices, \"Am I right?\"*\n\n**`/flip`** | `/alt` -- Identify the approach you'd take by default and state it. Then propose an alternative that uses a different mechanism (different data structure, different layer, different abstraction). State the conditions under which the alternative beats the default. Override: `/flip 3` for top 3 alternatives.\n*Auto-triggers on: architecture decisions where the \"easy\" answer may break at scale.*\n\n**`/assumptions`** | `/presume` -- Before answering, list every implicit assumption in the question/task. Then answer with assumptions explicit. The assumption list is often more valuable than the answer.\n*Auto-triggers on: architecture reviews, ambiguous requirements.*\n\n**`/premortem`** | `/postmortem` -- Assume the decision/project has already failed. Work backwards: what caused the failure? List 3-5 failure modes by likelihood. Focus on systemic risks, not edge cases.\n\n**`/tensions`** | `/perspectives` -- Answer from two named opposing perspectives (e.g., security engineer vs. shipping PM). Focus output on where they *disagree* -- that's where the real insight lives. Override roles: `/tensions [devops, security]`.\n\n## Combos\n\n**`/analyze`** = `/think` + `/edge` + `/verify-think` -- Code reviews, architecture, security-sensitive work. Synthesize findings into a unified recommendation -- don't just concatenate pattern outputs.\n*Auto-triggers on: code review requests.*\n\n**`/trade`** = `/confidence` + `/adversarial` + `/edge` -- Trade ideas, position analysis, market thesis.\n*Auto-triggers on: trade/position discussions.*\n\n## Conventions\n\n- Separate combined pattern outputs with `---`\n- Keep core answer prominent -- patterns enhance, not bury the response\n- Accept new pattern definitions mid-conversation (\"Add `/eli5` for explain like I'm 5\") -- apply for the session\n\n## Verify\n\n- Pattern marker present in output (e.g., `VERIFIED ANSWER:` for /verify)\n- Core answer remains prominent -- meta-reasoning enhances, doesn't bury it","tags":["meta","prompting","skills","iliaal","agent-skills","ai-coding-assistant","ai-tools","claude-code"],"capabilities":["skill","source-iliaal","skill-meta-prompting","topic-agent-skills","topic-ai-coding-assistant","topic-ai-tools","topic-claude-code","topic-skills"],"categories":["ai-skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/iliaal/ai-skills/meta-prompting","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add iliaal/ai-skills","source_repo":"https://github.com/iliaal/ai-skills","install_from":"skills.sh"}},"qualityScore":"0.456","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 13 github stars · SKILL.md body (4,248 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":"2026-05-18T19:07:02.584Z","embedding":null,"createdAt":"2026-05-09T01:05:35.609Z","updatedAt":"2026-05-18T19:07:02.584Z","lastSeenAt":"2026-05-18T19:07:02.584Z","tsv":"'-10':211 '-100':188 '-5':409 '-84':201 '/adversarial':21,79,486 '/alt':300 '/analyze':454 '/argue':80 '/assumptions':359 '/break':215 '/budget':147 '/check':243 '/commands':10 '/compare':165,179 '/conf':183 '/confidence':182,485 '/constrain':99,112 '/deep':148 '/edge':214,456,487 '/eli5':524 '/flip':299,339 '/format':116 '/ia-verify':276 '/json':115,142 '/perspectives':422 '/postmortem':394 '/premortem':393 '/presume':360 '/show':50 '/strict':100 '/tensions':421,450 '/think':49,65,455 '/trade':484 '/verify':20,545 '/verify-think':242,457 '/vs':166 '0':187 '1':210,246 '2':250 '3':87,103,252,258,340,343,408 '5':113,216,530 '500':152 '70':191,200,203 '85':135,198 'abstract':327 'accept':516 'accepted/rejected':63 'accuraci':174 'add':523 'alreadi':399 'altern':59,315,334,344 'ambigu':391 'analysi':131,491 'answer':43,45,82,164,247,269,272,354,362,371,384,423,508,543,547 'appli':31,33,531 'approach':221,303 'architectur':292,349,389,460 'ask':127 'assum':395 'assumpt':98,205,366,373,376 'auth':227 'auto':23,235,289,346,386,478,495 'auto-trigg':22,234,288,345,385,477,494 'backward':402 'beat':335 'blind':94 'block':122 'break':219,356 'buri':513,556 'bypass':228 'case':86,420 'caus':404 'challeng':251 'choic':295 'cite':106 'claim':186 'clear':161 'code':121,222,458,481 'combin':15,502 'combo':453 'command':14,278 'compar':167 'complex':176 'concaten':474 'concurr':224 'condit':231,330 'confid':46,133,213 'consid':60 'constraint':102 'context':26,241 'convent':500 'convers':522 'core':507,546 'cost':175 'could':72,255 'counterargu':89 'critic':294 'custom':140,178 'd':305 'data':322 'dead':155 'decis':57,293,350 'decision/project':397 'default':128,171,308,337 'definit':519 'depend':233 'devop':451 'differ':319,321,324,326 'dim1':180 'dim2':181 'dimens':172 'direct':248 'disagre':440 'discuss':499 'distinct':273 'doesn':554 'doubt':66 'e.g':41,429,541 'easi':353 'edg':419 'encod':226 'end':156 'engin':431 'enhanc':7,511,553 'everi':364 'explain':526 'explicit':374 'extend':149 'fail':400 'failur':406,410 'final':163,266 'find':466 'flag':70,189 'focus':414,435 'full':282 'group':194 'hedg':109 'high':197 'idea':489 'identifi':93,301 'implicit':365 'includ':204 'inlin':36,111 'inputs/scenarios':217 'insight':446 'investig':260 'json':120,130 'keep':506 'key':141,143 'languag':13 'layer':325 'left':17 'left-to-right':16 'like':527 'likelihood':413 'limit':138 'list':363,377,407 'live':447 'low':202 'm':529 'made':206 'mainten':177 'mark':38,265 'marker':537 'market':230,492 'max':105 'may':355 'mechan':320 'medium':199 'meta':2,5,551 'meta-prompt':1,4 'meta-reason':550 'methodolog':136 'mid':521 'mid-convers':520 'mode':411 'name':426 'natur':12 'need':264 'new':517 'note':28 'null/empty':223 'often':379 'oppos':85,427 'option':168 'output':32,436,476,504,540 'overflow':225 'overrid':110,338,448 'pars':240 'pattern':30,35,48,475,503,510,518,536 'perspect':428 'phase':245 'pipelin':287 'pivot':159 'pm':434 'point':58 'posit':490 'pr':285 'pre':284 'pre-pr':283 'present':538 'proceed':78 'promin':509,549 'prompt':3,6 'propos':313 'prose':125 'question/task':369 'rank':90 'rate':184,208 'real':445 'reason':8,52,158,552 'recommend':146,470 'remain':548 'request':483 'requir':392 'respond':117 'respons':249,515 'result':40 'review':390,459,482 'revis':44,271 'right':19,298 'risk':145,417 'role':449 'run':280 'scale':358 'schema':129 'score':134 'secur':238,430,452,462 'security-sensit':461 'sensit':463 'sentenc':104,114 'separ':162,501 'session':534 'sever':92 'ship':433 'show':51,154 'skill' 'skill-meta-prompting' 'slash':277 'sourc':107 'source-iliaal' 'space':151 'specul':193 'speed':173 'spot':95 'state':310,328 'steelman':83 'step':54,56,69 'step-by-step':53 'strategi':229 'string':132,137,139 'strongest':88 'structur':323 'summari':144 'surround':124 'synthes':465 'system':416 'tabl':170 'take':306 'thesi':493 'think':150 'three':244 'tier':47,196 'tight':101 'time':232 'top':342 'topic-agent-skills' 'topic-ai-coding-assistant' 'topic-ai-tools' 'topic-claude-code' 'topic-skills' 'trade':488 'trade/position':498 'trigger':24,236,290,347,387,479,496 'two':425 'unifi':469 'unless':126 'unstat':97 'updat':262 'use':317 'valid':119,239 'valuabl':381 'verif':286 'verifi':42,259,268,535,542 'via':9 'vs':432 'warrant':27 'way':253 'word':153 'work':401,464 'wrong':74,257","prices":[{"id":"9173f9cc-408b-4b59-b13c-5ef5bbc8e476","listingId":"59887f75-b072-4750-b600-1544c01ae5ed","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"iliaal","category":"ai-skills","install_from":"skills.sh"},"createdAt":"2026-05-09T01:05:35.609Z"}],"sources":[{"listingId":"59887f75-b072-4750-b600-1544c01ae5ed","source":"github","sourceId":"iliaal/ai-skills/meta-prompting","sourceUrl":"https://github.com/iliaal/ai-skills/tree/master/skills/meta-prompting","isPrimary":false,"firstSeenAt":"2026-05-09T01:05:35.609Z","lastSeenAt":"2026-05-18T19:07:02.584Z"}],"details":{"listingId":"59887f75-b072-4750-b600-1544c01ae5ed","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"iliaal","slug":"meta-prompting","github":{"repo":"iliaal/ai-skills","stars":13,"topics":["agent-skills","ai-coding-assistant","ai-tools","claude-code","skills"],"license":"mit","html_url":"https://github.com/iliaal/ai-skills","pushed_at":"2026-05-16T13:15:17Z","description":"Curated collection of agent skills for AI coding assistants.","skill_md_sha":"a4fd59641bed71e96379eefa7cc114ee3a491b32","skill_md_path":"skills/meta-prompting/SKILL.md","default_branch":"master","skill_tree_url":"https://github.com/iliaal/ai-skills/tree/master/skills/meta-prompting"},"layout":"multi","source":"github","category":"ai-skills","frontmatter":{"name":"meta-prompting","description":">-"},"skills_sh_url":"https://skills.sh/iliaal/ai-skills/meta-prompting"},"updatedAt":"2026-05-18T19:07:02.584Z"}}