[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71NDFZPXWQJ8W32KEH4Q4E7":3,"skill-same:01M71NDFZPXWQJ8W32KEH4Q4E7":74},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":51,"repositoryHomepageUrl":56,"ownerHtmlUrl":57,"ownerBlogUrl":58,"ownerDescription":6,"updatedAt":59,"install":60,"resources":73},"01M71NDFZPXWQJ8W32KEH4Q4E7","model-qa",null,"Use when an org role acts as model QA specialist and must independently audit an ML or statistical model: replication, calibration, drift, fairness, governance. Rates findings by severity with PSI, SHAP and champion-challenger tests; audits models others built.","monoes\u002Fmonomind","monoes","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F276744696?v=4","01M71KWF37AJ2NJBP3KWBVSDGT","01M71KWF4CSF8ADH0M6HA09VD2",23,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fmodel-qa\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"qa","model qa","model","governance","audit","github","monomind","with","shap","drift","query","ml","PSI","when","use","impact","tests","replication","severity","as","2026-11-10 10:56:23",false,"---\nname: model-qa\ndescription: \"Use when an org role acts as model QA specialist and must independently audit an ML or statistical model: replication, calibration, drift, fairness, governance. Rates findings by severity with PSI, SHAP and champion-challenger tests; audits models others built.\"\ntags: [\"ai-ml\",\"engineering\",\"audit\",\"evaluation\"]\ntools: [\"monograph_query\",\"monograph_context\",\"monograph_impact\"]\nlicense: Apache-2.0\nsource: https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\n---\n# Model QA Specialist — Best Practices\n\n## Focus\nIndependently audits ML\u002Fstatistical models end-to-end — documentation, data, replication, calibration, and fairness — to certify whether a model is sound before or during production use.\n\n## Best practices\n- Never audit a model you built yourself — independence is the whole point; findings from a self-review are structurally suspect.\n- Require full reproducibility: every analysis must run from raw data to final output via a versioned, self-contained script — no manual steps.\n- Replicate the model from documented methodology and compare outputs against the original (parameter deltas, score distributions) before trusting either.\n- Test calibration explicitly, not just discrimination — a model can rank correctly (good AUC) while its predicted probabilities are badly miscalibrated.\n- Check feature and population stability over time (PSI or similar) — a model that was sound at training time can silently drift out of validity.\n- Rate every finding by severity (High\u002FMedium\u002FLow\u002FInfo) with quantified business impact — \"this seems off\" is not a finding.\n- Audit fairness across protected\u002Fsegment groups explicitly, not just aggregate performance.\n- Verify governance basics — model inventory, approval trail, monitoring plan — exist and are current, before diving into the statistics.\n\n## Common pitfalls\n- Accepting in-sample or training metrics as sufficient evidence of production soundness, skipping out-of-time validation.\n- Declaring \"the model is wrong\" without quantifying the impact or proposing a remediation.\n- Checking aggregate performance only, missing a segment where the model fails badly but is diluted in the overall number.\n- Treating a passed discrimination test (AUC\u002FGini) as sufficient without also checking calibration — the two measure different things and both can fail independently.\n- Skipping documentation\u002Fgovernance review because the statistics \"look fine\" — an undocumented or unapproved model is a finding regardless of accuracy.\n\n## Tools & techniques\n- Population Stability Index (PSI) per feature and per period to quantify distribution drift.\n- Hosmer-Lemeshow test, Brier score, and reliability diagrams for calibration validation.\n- SHAP global (beeswarm\u002Fimportance) and local (waterfall) analysis to verify learned relationships match documented rationale.\n- Partial Dependence Plots (including 2D interaction plots) to check for expected monotonic\u002Fdirectional relationships and detect learned interaction effects.\n- Champion-challenger benchmarking with statistical significance testing (e.g. DeLong test for AUC differences) before recommending a model swap.\n","# Model QA Specialist — Best Practices\n\n## Focus\nIndependently audits ML\u002Fstatistical models end-to-end — documentation, data, replication, calibration, and fairness — to certify whether a model is sound before or during production use.\n\n## Best practices\n- Never audit a model you built yourself — independence is the whole point; findings from a self-review are structurally suspect.\n- Require full reproducibility: every analysis must run from raw data to final output via a versioned, self-contained script — no manual steps.\n- Replicate the model from documented methodology and compare outputs against the original (parameter deltas, score distributions) before trusting either.\n- Test calibration explicitly, not just discrimination — a model can rank correctly (good AUC) while its predicted probabilities are badly miscalibrated.\n- Check feature and population stability over time (PSI or similar) — a model that was sound at training time can silently drift out of validity.\n- Rate every finding by severity (High\u002FMedium\u002FLow\u002FInfo) with quantified business impact — \"this seems off\" is not a finding.\n- Audit fairness across protected\u002Fsegment groups explicitly, not just aggregate performance.\n- Verify governance basics — model inventory, approval trail, monitoring plan — exist and are current, before diving into the statistics.\n\n## Common pitfalls\n- Accepting in-sample or training metrics as sufficient evidence of production soundness, skipping out-of-time validation.\n- Declaring \"the model is wrong\" without quantifying the impact or proposing a remediation.\n- Checking aggregate performance only, missing a segment where the model fails badly but is diluted in the overall number.\n- Treating a passed discrimination test (AUC\u002FGini) as sufficient without also checking calibration — the two measure different things and both can fail independently.\n- Skipping documentation\u002Fgovernance review because the statistics \"look fine\" — an undocumented or unapproved model is a finding regardless of accuracy.\n\n## Tools & techniques\n- Population Stability Index (PSI) per feature and per period to quantify distribution drift.\n- Hosmer-Lemeshow test, Brier score, and reliability diagrams for calibration validation.\n- SHAP global (beeswarm\u002Fimportance) and local (waterfall) analysis to verify learned relationships match documented rationale.\n- Partial Dependence Plots (including 2D interaction plots) to check for expected monotonic\u002Fdirectional relationships and detect learned interaction effects.\n- Champion-challenger benchmarking with statistical significance testing (e.g. DeLong test for AUC differences) before recommending a model swap.\n",{"name":5,"tags":43,"tools":47,"source":51,"license":14,"description":7},[44,45,22,46],"ai-ml","engineering","evaluation",[48,49,50],"monograph_query","monograph_context","monograph_impact","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind","main","f7cd4595d0f8959822163e14f9787117765c931a","packages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fmodel-qa",2,"https:\u002F\u002Fmonoes.github.io\u002Fmonomind\u002F","https:\u002F\u002Fgithub.com\u002Fmonoes","https:\u002F\u002Fmonoes.me","2026-11-10 01:53:48",{"source":8,"skillName":5,"commands":61,"prompt":71,"zipAvailable":72},[62,65,68],{"tool":63,"command":64},"npx","npx skills add monoes\u002Fmonomind --skill model-qa",{"tool":66,"command":67},"bunx","bunx skills add monoes\u002Fmonomind --skill model-qa",{"tool":69,"command":70},"pnpm","pnpm dlx skills add monoes\u002Fmonomind --skill model-qa","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fmodel-qa\u002FSKILL.md",true,[],[75,100,136,162,185,212],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":99},"01M71PB3SHWKMNJM6MD65VX8D6","zk-steward","Org role guidance for a ZK steward: engineer zero-knowledge proof systems with circuit design, SNARK vs STARK choice and security review. This is zero-knowledge cryptography, not Zettelkasten note-taking.","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fzk-steward\u002FSKILL.md",[],[82,83,84,85,86,87,88,89,33,28,90,91,24,92,93,94,95,96,97,98],"zk","zk steward","steward","zero-knowledge","engineer","choice","design","knowledge","snark","review","security review","stark","zettelkasten","cryptography","security","org","blockchain","2026-11-10 13:02:02",{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":104,"licenseFileUrl":105,"sourceHtmlUrl":106,"categories":107,"keywords":118,"publishedAt":99},"01M71PB18SDG28WBKPKJNB814K","writing-plans","Use when you have a spec or requirements for a multi-step coding task and must write an implementation plan before touching code. Produces bite-sized TDD tasks with exact file paths, code and commands for an engineer with zero context.","MIT","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fwriting-plans\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fwriting-plans\u002FSKILL.md",[108,113],{"categoryNo":109,"categoryName":110,"taxonomyType":111,"taxonomyTypeDesc":112},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":114,"categoryName":115,"taxonomyType":116,"taxonomyTypeDesc":117},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[119,120,121,86,122,123,31,124,28,125,126,127,128,129,130,131,132,133,134,135],"writing","writing plans","plans","superpowers","implementation plan","you","code","tasks","plan","product","write","commands","paths","suggest","coding","context","ZeRO",{"skillNo":137,"name":138,"nameZh":6,"description":139,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":140,"sourceHtmlUrl":141,"categories":142,"keywords":146,"publishedAt":99},"01M71PAXFM488X9ARXK0T04TF5","write-spec","Write a feature spec or PRD from a problem statement or feature idea. Use when turning a vague idea or user request into a structured document, scoping a feature with goals and non-goals, defining success metrics and acceptance criteria, or breaking a big ask into a phased spec.","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fwrite-spec\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fwrite-spec\u002FSKILL.md",[143],{"categoryNo":144,"categoryName":145,"taxonomyType":116,"taxonomyTypeDesc":117},"01JBSEEDCAT000000000000031","产品经理",[129,147,148,149,150,89,151,31,152,153,154,128,155,156,157,158,159,160,32,161],"write spec","spec","user request","problem","acceptance","ask","vague","work","request","scoping","idea","metrics","feature","problem statement","document",{"skillNo":163,"name":164,"nameZh":6,"description":165,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":166,"sourceHtmlUrl":167,"categories":168,"keywords":172,"publishedAt":99},"01M71PASSZY800C0RPEARBS918","write-query","Write optimized SQL for your dialect with best practices. Use when translating a natural-language data need into SQL, building a multi-CTE query with joins and aggregations, optimizing a query against a large partitioned table, or getting dialect-specific syntax for Snowflake, BigQuery, Postgres, etc.","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fwrite-query\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fwrite-query\u002FSKILL.md",[169],{"categoryNo":170,"categoryName":171,"taxonomyType":116,"taxonomyTypeDesc":117},"01JBSEEDCAT000000000000033","数据分析师",[28,129,173,174,175,176,89,32,31,177,178,25,179,154,23,180,181,182,183,184],"write query","postgres","for","snowflake","best practices","natural-language","multi","sql","https","syntax","plugins","data",{"skillNo":186,"name":187,"nameZh":6,"description":188,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":189,"categories":190,"keywords":197,"publishedAt":99},"01M71PAPTX0T3PF6ETFVR96SJJ","workflow-automation","Org role guidance for workflow automation: build and maintain CI\u002FCD definitions in GitHub Actions or equivalent: triggers, jobs, matrices, caching. Covers path filters, dependency caching, parallel jobs, timeouts and fail-fast ordering.","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fworkflow-automation\u002FSKILL.md",[191,194],{"categoryNo":192,"categoryName":193,"taxonomyType":111,"taxonomyTypeDesc":112},"01JBSEEDCAT000000000000011","运维与部署",{"categoryNo":195,"categoryName":196,"taxonomyType":111,"taxonomyTypeDesc":112},"01JBSEEDCAT000000000000023","Skill与智能体开发",[198,199,200,201,175,202,203,204,205,33,206,207,208,28,24,23,209,210,211,181],"Workflow Automation","automation","workflow","parallel","actions","ci","triggers","github actions","maintain","filters","caching","build","dependency","devops",{"skillNo":213,"name":214,"nameZh":6,"description":215,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":216,"categories":217,"keywords":219,"publishedAt":99},"01M71PAM19Z1XTV9NPCFSE2MY8","workflow-architect","Org role guidance for a workflow architect: map full workflow trees for systems, user journeys and agents, covering every branch and failure. Covers discovering workflows from code, handoff contracts, timeouts, recovery paths and observable state.","https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fworkflow-architect\u002FSKILL.md",[218],{"categoryNo":195,"categoryName":196,"taxonomyType":111,"taxonomyTypeDesc":112},[200,220,221,222,223,224,225,175,226,227,125,228,229,230,24,23,231,232,181,233],"workflow architect","architect","trees","user journeys","branch","planning","map","workflows","architecture","state","leadership","recovery","operations","systems"]