[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6AE2RSGGP3KA22ZHW6KHNEG":3,"skill-same:01M6AE2RSGGP3KA22ZHW6KHNEG":68},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":49,"ownerHtmlUrl":50,"ownerBlogUrl":51,"ownerDescription":52,"updatedAt":53,"install":54,"resources":67},"01M6AE2RSGGP3KA22ZHW6KHNEG","qdrant-search-quality-diagnosis",null,"Diagnoses Qdrant search quality issues. Use when someone reports 'results are bad', 'wrong results', 'not relevant results', 'missing matches', 'recall is low', 'approximate search worse than exact', 'which embedding model', or 'quality dropped after quantization'. Also use when search quality degrades without obvious changes.","github\u002Fawesome-copilot","github","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F9919?v=4","01M6AC83DS133TPXFSP1DHNCC6","GitHub","01M6AD3PBDYPK7NVFJNJ66Y8RS",41046,"MIT","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fqdrant-search-quality\u002Fdiagnosis\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"search","Qdrant","qdrant search quality","qdrant search quality diagnosis","quality","diagnosis","when","issues","changes","which","relevant","embedding","Quantization","missing","bad","reports","use","results","model","recall","2026-11-10 10:12:30",false,"---\nname: qdrant-search-quality-diagnosis\ndescription: \"Diagnoses Qdrant search quality issues. Use when someone reports 'results are bad', 'wrong results', 'not relevant results', 'missing matches', 'recall is low', 'approximate search worse than exact', 'which embedding model', or 'quality dropped after quantization'. Also use when search quality degrades without obvious changes.\"\n---\n\n# How to Diagnose Bad Search Quality\n\nBefore tuning, establish baselines. Use exact KNN as ground truth, compare against approximate HNSW. Target >95% recall@K for production.\n\n## Don't Know What's Wrong Yet\n\nUse when: results are irrelevant or missing expected matches and you need to isolate the cause.\n\n- Test with `exact=true` to bypass HNSW approximation [Search API](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Ftutorials-search-engineering\u002Fretrieval-quality\u002F?s=standard-mode-vs-exact-search)\n- Exact search bad = model or search pipeline problem. Exact good, approximate bad = tune HNSW.\n- Check if quantization degrades quality (compare with and without)\n- Check if filters are too restrictive (then you might need to use ACORN)\n- If duplicate results from chunked documents, use Grouping API to deduplicate [Grouping](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=grouping-api)\n\nPayload filtering and sparse vector search are different things. Metadata (dates, categories, tags) goes in payload for filtering. Text content goes in sparse vectors for search.\n\n## Approximate Search Worse Than Exact\n\nUse when: exact search returns good results but HNSW approximation misses them.\n\n- Increase `hnsw_ef` at query time [Search params](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Foperations\u002Foptimize\u002F?s=fine-tuning-search-parameters)\n- Increase `ef_construct` (200+ for high quality) [HNSW config](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index)\n- Increase `m` (16 default, 32 for high recall) [HNSW config](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index)\n- Enable oversampling + rescore with quantization [Search with quantization](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F?s=searching-with-quantization)\n- ACORN for filtered queries (v1.16+) [ACORN](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=acorn-search-algorithm)\n\nBinary quantization requires rescore. Without it, quality loss is severe. Use oversampling (3-5x minimum for binary) to recover recall. Always test quantization impact on your data before production. [Quantization](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F)\n\n## Wrong Embedding Model\n\nUse when: exact search also returns bad results.\n\nTest top 3 MTEB models on 100-1000 sample queries, measure recall@10. Domain-specific models often outperform general models. [Hosted inference](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Finference\u002F)\n\n## Unoptimized Search Pipeline\n\nUse when: exact search also returns bad results and model choice is confirmed by user.\n\nOptimize search according to advanced search-strategies skill.\n\n## What NOT to Do\n\n- Tune Qdrant before verifying the model is right for the task (most quality issues are model issues)\n- Use binary quantization without rescore (severe quality loss)\n- Set `hnsw_ef` lower than results requested (guaranteed bad recall)\n- Skip payload indexes on filtered fields then blame quality (HNSW can't traverse filtered-out nodes, and filterable HNSW is built only if payload indexes were set up prior)\n- Deploy without baseline recall or other search relevance metrics (no way to measure regressions)\n- Confuse payload filtering with sparse vector search (different things, different config)\n","# How to Diagnose Bad Search Quality\n\nBefore tuning, establish baselines. Use exact KNN as ground truth, compare against approximate HNSW. Target >95% recall@K for production.\n\n## Don't Know What's Wrong Yet\n\nUse when: results are irrelevant or missing expected matches and you need to isolate the cause.\n\n- Test with `exact=true` to bypass HNSW approximation [Search API](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Ftutorials-search-engineering\u002Fretrieval-quality\u002F?s=standard-mode-vs-exact-search)\n- Exact search bad = model or search pipeline problem. Exact good, approximate bad = tune HNSW.\n- Check if quantization degrades quality (compare with and without)\n- Check if filters are too restrictive (then you might need to use ACORN)\n- If duplicate results from chunked documents, use Grouping API to deduplicate [Grouping](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=grouping-api)\n\nPayload filtering and sparse vector search are different things. Metadata (dates, categories, tags) goes in payload for filtering. Text content goes in sparse vectors for search.\n\n## Approximate Search Worse Than Exact\n\nUse when: exact search returns good results but HNSW approximation misses them.\n\n- Increase `hnsw_ef` at query time [Search params](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Foperations\u002Foptimize\u002F?s=fine-tuning-search-parameters)\n- Increase `ef_construct` (200+ for high quality) [HNSW config](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index)\n- Increase `m` (16 default, 32 for high recall) [HNSW config](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Findexing\u002F?s=vector-index)\n- Enable oversampling + rescore with quantization [Search with quantization](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F?s=searching-with-quantization)\n- ACORN for filtered queries (v1.16+) [ACORN](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fsearch\u002Fsearch\u002F?s=acorn-search-algorithm)\n\nBinary quantization requires rescore. Without it, quality loss is severe. Use oversampling (3-5x minimum for binary) to recover recall. Always test quantization impact on your data before production. [Quantization](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Fmanage-data\u002Fquantization\u002F)\n\n## Wrong Embedding Model\n\nUse when: exact search also returns bad results.\n\nTest top 3 MTEB models on 100-1000 sample queries, measure recall@10. Domain-specific models often outperform general models. [Hosted inference](https:\u002F\u002Fsearch.qdrant.tech\u002Fmd\u002Fdocumentation\u002Finference\u002F)\n\n## Unoptimized Search Pipeline\n\nUse when: exact search also returns bad results and model choice is confirmed by user.\n\nOptimize search according to advanced search-strategies skill.\n\n## What NOT to Do\n\n- Tune Qdrant before verifying the model is right for the task (most quality issues are model issues)\n- Use binary quantization without rescore (severe quality loss)\n- Set `hnsw_ef` lower than results requested (guaranteed bad recall)\n- Skip payload indexes on filtered fields then blame quality (HNSW can't traverse filtered-out nodes, and filterable HNSW is built only if payload indexes were set up prior)\n- Deploy without baseline recall or other search relevance metrics (no way to measure regressions)\n- Confuse payload filtering with sparse vector search (different things, different config)\n",{"name":5,"description":7},"main","5269f7ea2d6c3a9bfda5ed878b6726cf509121fd","skills\u002Fqdrant-search-quality\u002Fdiagnosis",5340,"https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot","https:\u002F\u002Fawesome-copilot.github.com\u002F","https:\u002F\u002Fgithub.com\u002Fgithub","https:\u002F\u002Fgithub.com\u002Fabout","How people build software.","2026-11-11 13:41:32",{"source":8,"skillName":5,"commands":55,"prompt":65,"zipAvailable":66},[56,59,62],{"tool":57,"command":58},"npx","npx skills add github\u002Fawesome-copilot --skill qdrant-search-quality-diagnosis",{"tool":60,"command":61},"bunx","bunx skills add github\u002Fawesome-copilot --skill qdrant-search-quality-diagnosis",{"tool":63,"command":64},"pnpm","pnpm dlx skills add github\u002Fawesome-copilot --skill qdrant-search-quality-diagnosis","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fqdrant-search-quality\u002Fdiagnosis\u002FSKILL.md",true,[],[69,102,130,160,193,210],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":83,"publishedAt":39},"01M70KXZNX9ES38N2A8M58ZZMZ","supabase-rls-policies","Write, review, and test Supabase Row Level Security (RLS) policies so that one user cannot read or change another user's rows. Use when creating or changing tables, policies, views, or security definer functions in a Supabase project, when reviewing a migration that touches RLS, or when asked to prove user isolation with pgTAP tests that sign in as two users.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fsupabase-rls-policies\u002FSKILL.md",[75,80],{"categoryNo":76,"categoryName":77,"taxonomyType":78,"taxonomyTypeDesc":79},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":81,"categoryName":82,"taxonomyType":78,"taxonomyTypeDesc":79},"01JBSEEDCAT000000000000023","Skill与智能体开发",[84,85,86,87,88,25,89,90,91,92,93,94,95,96,97,98,35,99,100,101],"supabase","rls","supabase rls","supabase rls policies","project","review","change","read","write","security","isolation","tables","as","tests","prove","users","functions","with",{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":109,"publishedAt":39},"01M70KWA5YWXDV6V8DZ19E3T2Y","skill-distribution-audit","Audit a skill collection before release or investigate installer omissions, archived\u002Ftemplate leakage, duplicate identities, and manifest drift. Reconcile the intended public source set, distribution manifest, and actual discovery output for a specified consumer version. Includes a deterministic set-comparison helper.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fskill-distribution-audit\u002FSKILL.md",[108],{"categoryNo":81,"categoryName":82,"taxonomyType":78,"taxonomyTypeDesc":79},[110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129],"distribution","skill distribution audit","audit","SET","deterministic","installer","for","discovery","comparison","helper","manifest","version","collection","drift","output","investigate","duplicate","template","release","reconcile",{"skillNo":131,"name":132,"nameZh":6,"description":133,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":134,"sourceHtmlUrl":135,"categories":136,"keywords":142,"publishedAt":39},"01M70KCJPFVW7PPQ49CC5R8JGK","logicapps-custom-code-net10-migration","Migrate Azure Logic Apps Standard local-function custom-code projects from .NET 8 to .NET 10. Use for legacy Logic Apps custom-code workspace upgrades, project\u002Fpackage configuration, custom-worker settings, publication, and integrated VS Code debugging. Not for ordinary .NET applications, standalone Azure Functions, Consumption workflows, or cloud deployment.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Flogicapps-custom-code-net10-migration\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Flogicapps-custom-code-net10-migration\u002FSKILL.md",[137],{"categoryNo":138,"categoryName":139,"taxonomyType":140,"taxonomyTypeDesc":141},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[143,144,145,146,147,148,116,88,149,150,151,152,153,154,155,35,156,157,158,159],"custom","code","logicapps custom code net10 migration","migration","projects","standalone","deployment","applications","local","azure","worker","upgrades","package","apps","Legacy","debugging","workflows",{"skillNo":161,"name":162,"nameZh":6,"description":163,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":164,"sourceHtmlUrl":165,"categories":166,"keywords":171,"publishedAt":39},"01M6AEBN37RQMXJ3298PEY7PG1","winmd-api-search","Find and explore Windows desktop APIs. Use when building features that need platform capabilities — camera, file access, notifications, UI controls, AI\u002FML, sensors, networking, etc. Discovers the right API for a task and retrieves full type details (methods, properties, events, enumeration values).","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fwinmd-api-search\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fwinmd-api-search\u002FSKILL.md",[167,170],{"categoryNo":168,"categoryName":169,"taxonomyType":78,"taxonomyTypeDesc":79},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":138,"categoryName":139,"taxonomyType":140,"taxonomyTypeDesc":141},[172,19,173,174,175,176,177,178,179,180,181,182,183,184,185,25,186,187,188,189,190,116,191,192],"winmd api search","api","windows","desktop","file access","enumeration","ui","ml","file","networking","capabilities","notifications","explore","access","features","full","camera","events","platform","methods","controls",{"skillNo":194,"name":195,"nameZh":6,"description":196,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":197,"licenseFileUrl":198,"sourceHtmlUrl":199,"categories":200,"keywords":206,"publishedAt":39},"01M6AE571R3SC1ZGE202B3NB5S","resemble-detect","Deepfake detection and media safety — detect AI-generated audio, images, video, and text, trace synthesis sources, apply watermarks, verify speaker identity, and analyze media intelligence using Resemble AI","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fresemble-detect\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fresemble-detect\u002FSKILL.md",[201,202,205],{"categoryNo":168,"categoryName":169,"taxonomyType":78,"taxonomyTypeDesc":79},{"categoryNo":203,"categoryName":204,"taxonomyType":78,"taxonomyTypeDesc":79},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":81,"categoryName":82,"taxonomyType":78,"taxonomyTypeDesc":79},[207,113,180,208,209],"resemble detect","https","identity",{"skillNo":211,"name":212,"nameZh":6,"description":213,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":197,"licenseFileUrl":214,"sourceHtmlUrl":215,"categories":216,"keywords":221,"publishedAt":39},"01M6AE30BSK2HCY4TQQ22Z8JRD","quality-playbook","Run a complete quality engineering audit on any codebase. Derives behavioral requirements from the code, generates spec-traced functional tests, runs a three-pass code review with regression tests, executes a multi-model spec audit (Council of Three), and produces a consolidated bug report with TDD-verified patches. Finds the 35% of real defects that structural code review alone cannot catch. Works with any language. Trigger on 'quality playbook', 'spec audit', 'Council of Three', 'fitness-to-purpose', or 'coverage theater'.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fquality-playbook\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fquality-playbook\u002FSKILL.md",[217,220],{"categoryNo":218,"categoryName":219,"taxonomyType":78,"taxonomyTypeDesc":79},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":81,"categoryName":82,"taxonomyType":78,"taxonomyTypeDesc":79},[222,223,23,224,144,89,9,112,225,226,208,227,228,229,230,231,232,233,234,235,236,237,97,238,37,239,240],"playbook","quality playbook","code review","tdd","run","report","spec","bug report","council","engineering","functional","spec audit","bug","verified","multi-model","three","coverage","regression","structural"]