[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EW9DBBWDZE65ZC7S0DXC59":3,"skill-same:01M7EW9DBBWDZE65ZC7S0DXC59":47},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M7EW9DBBWDZE65ZC7S0DXC59","vector-databases",null,"Choose and operate vector databases — indexing algorithms, filtering, scaling, and production concerns. Use when storing and querying embeddings at scale.","TanNhatCMS\u002Fjarvis","TanNhatCMS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113796420?v=4","01M753MV17KG5CD3152G3WQEQY","KietNT","01M7EKEDYVYQ6PWVJJMZ911VMX",0,"MIT","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-vector-databases\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: vector-databases\ndescription: Choose and operate vector databases — indexing algorithms, filtering, scaling, and production concerns. Use when storing and querying embeddings at scale.\ncategory: ai-research\n---\n\n# Vector Databases\n\nA vector database stores embeddings and finds nearest neighbors fast at scale. The decisions: which \nindex algorithm, how to filter, how to scale, and how to keep results good as data grows.\n\n## Overview\n\nCore operation: given a query vector, return the k most similar stored vectors — approximately \n(ANN) for speed at scale. HNSW graphs dominate for the quality\u002Fspeed balance; IVF partitions for \nhuge datasets. On top: metadata filtering (so \"similar\" also means \"in the right category\u002Fdate\"), \nhybrid search hooks, and operational concerns — indexing throughput, recall tuning, and \nmulti-tenancy.\n\n## When to use\n\n- RAG retrieval over large document collections.\n- Semantic search features in products.\n- Recommendation or deduplication at scale.\n- Anytime brute-force similarity search gets too slow.\n\n## Core concepts\n\n- **ANN indices**: HNSW (graph-based, excellent recall\u002Flatency, memory-hungry), IVF (partitioned, \nscales to billions, needs tuning), and others. HNSW is the default; IVF for the largest scales.\n- **Recall vs. speed**: ANN parameters trade accuracy for speed (HNSW's ef_search, IVF's nprobe). \nTune on your data: measure recall@k against brute force on a sample.\n- **Metadata filtering**: pre-filter (filter then search — precise, can be slow), post-filter \n(search then filter — fast, may return too few), or index-aware hybrid. Your filter selectivity \ndecides.\n- **Namespaces\u002Ftenancy**: partition by tenant, project, or corpus — for isolation, deletion, and \nper-tenant tuning. Design tenancy before data grows.\n- **Distance metrics**: cosine, dot product, Euclidean — must match how embeddings were produced. \nSet once, consistently.\n- **Freshness**: upserts, deletes, and index rebuilds. Know your system's update latency and \nwhether deletes are real or tombstoned.\n\n## Practical workflow\n\n1. Define the workload: corpus size, query rate, filter patterns, latency budget, recall target.\n2. Choose the index: HNSW for most; IVF-family for 100M+ scale. Start managed unless you have ops \ncapacity.\n3. Set the distance metric to match your embeddings; normalize consistently.\n4. Design metadata schema and tenancy up front — retrofitting partitions is painful.\n5. Tune ANN parameters on your data: sweep ef_search\u002Fnprobe, plot recall vs. latency, pick the knee.\n6. Load-test with production-like queries and filters; monitor recall drift as the corpus grows.\n\n```text\nSelection checklist:\n[ ] Corpus size now + 12-month projection\n[ ] p95 latency budget defined\n[ ] Recall@k target (measured vs brute force)\n[ ] Filter patterns identified (selectivity known)\n[ ] Tenancy\u002Fisolation requirements set\n[ ] Update\u002Fdelete latency acceptable\n```\n\n## Common pitfalls\n\n- **Untuned ANN params**: default ef_search giving 80% recall when you need 95%. Tune on your data.\n- **Metric mismatch**: cosine index with dot-product-trained embeddings. Consistent end-to-end or \nbroken ranking.\n- **Post-filter starvation**: heavy filtering after ANN search returns fewer than k results. Use \npre-filtering for selective filters.\n- **No tenancy plan**: one flat index for all customers, then a deletion\u002Fisolation requirement \narrives. Partition early.\n- **Ignoring write path**: great search, terrible ingestion — or deletes that don't delete. Test \nthe full lifecycle.\n- **Recall measured once**: index quality degrades as data distribution shifts. Re-measure \nperiodically.\n","# Vector Databases\n\nA vector database stores embeddings and finds nearest neighbors fast at scale. The decisions: which \nindex algorithm, how to filter, how to scale, and how to keep results good as data grows.\n\n## Overview\n\nCore operation: given a query vector, return the k most similar stored vectors — approximately \n(ANN) for speed at scale. HNSW graphs dominate for the quality\u002Fspeed balance; IVF partitions for \nhuge datasets. On top: metadata filtering (so \"similar\" also means \"in the right category\u002Fdate\"), \nhybrid search hooks, and operational concerns — indexing throughput, recall tuning, and \nmulti-tenancy.\n\n## When to use\n\n- RAG retrieval over large document collections.\n- Semantic search features in products.\n- Recommendation or deduplication at scale.\n- Anytime brute-force similarity search gets too slow.\n\n## Core concepts\n\n- **ANN indices**: HNSW (graph-based, excellent recall\u002Flatency, memory-hungry), IVF (partitioned, \nscales to billions, needs tuning), and others. HNSW is the default; IVF for the largest scales.\n- **Recall vs. speed**: ANN parameters trade accuracy for speed (HNSW's ef_search, IVF's nprobe). \nTune on your data: measure recall@k against brute force on a sample.\n- **Metadata filtering**: pre-filter (filter then search — precise, can be slow), post-filter \n(search then filter — fast, may return too few), or index-aware hybrid. Your filter selectivity \ndecides.\n- **Namespaces\u002Ftenancy**: partition by tenant, project, or corpus — for isolation, deletion, and \nper-tenant tuning. Design tenancy before data grows.\n- **Distance metrics**: cosine, dot product, Euclidean — must match how embeddings were produced. \nSet once, consistently.\n- **Freshness**: upserts, deletes, and index rebuilds. Know your system's update latency and \nwhether deletes are real or tombstoned.\n\n## Practical workflow\n\n1. Define the workload: corpus size, query rate, filter patterns, latency budget, recall target.\n2. Choose the index: HNSW for most; IVF-family for 100M+ scale. Start managed unless you have ops \ncapacity.\n3. Set the distance metric to match your embeddings; normalize consistently.\n4. Design metadata schema and tenancy up front — retrofitting partitions is painful.\n5. Tune ANN parameters on your data: sweep ef_search\u002Fnprobe, plot recall vs. latency, pick the knee.\n6. Load-test with production-like queries and filters; monitor recall drift as the corpus grows.\n\n```text\nSelection checklist:\n[ ] Corpus size now + 12-month projection\n[ ] p95 latency budget defined\n[ ] Recall@k target (measured vs brute force)\n[ ] Filter patterns identified (selectivity known)\n[ ] Tenancy\u002Fisolation requirements set\n[ ] Update\u002Fdelete latency acceptable\n```\n\n## Common pitfalls\n\n- **Untuned ANN params**: default ef_search giving 80% recall when you need 95%. Tune on your data.\n- **Metric mismatch**: cosine index with dot-product-trained embeddings. Consistent end-to-end or \nbroken ranking.\n- **Post-filter starvation**: heavy filtering after ANN search returns fewer than k results. Use \npre-filtering for selective filters.\n- **No tenancy plan**: one flat index for all customers, then a deletion\u002Fisolation requirement \narrives. Partition early.\n- **Ignoring write path**: great search, terrible ingestion — or deletes that don't delete. Test \nthe full lifecycle.\n- **Recall measured once**: index quality degrades as data distribution shifts. Re-measure \nperiodically.\n",{"name":5,"category":24,"description":7},"ai-research","main","af388a77c2fe5e0b4fd879cca4300a0ffa9a6579","skills\u002Fams-vector-databases",1,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis","https:\u002F\u002Fgithub.com\u002FTanNhatCMS","https:\u002F\u002Fmrkiet.name.vn","2026-11-07 14:23:08",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add TanNhatCMS\u002Fjarvis --skill vector-databases",{"tool":39,"command":40},"bunx","bunx skills add TanNhatCMS\u002Fjarvis --skill vector-databases",{"tool":42,"command":43},"pnpm","pnpm dlx skills add TanNhatCMS\u002Fjarvis --skill vector-databases","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-vector-databases\u002FSKILL.md",true,[],[48,55,63,70,77,84],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7EWYJR0J8DARY7E59Z6Y9MV","web-scrape","Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open\u002Ffetch fails on JS-heavy pages, or when you need clean markdown\u002Fstructured extraction from a site. Requires FIRECRAWL_API_KEY.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fweb-scrape\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":59,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7EWYBN5P6F2AJG3FR24T9BP","design-taste-frontend","tasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curated gallery sites (One Page Love, Awwwards Sites of the Day). Deals researched directions (idea, composition, type, palette, photographs, section blueprints) from a library, enforces a strict no-slop house style with a real-browser inspector, and runs a look-and-fix loop on screenshots. Use this whenever you build, design, redesign, restyle, polish or review a landing page, marketing site, portfolio, product page, one-pager, hero or any public-facing web page, in any stack (HTML\u002FCSS, React, Next.js, Vue, Svelte, Astro, Tailwind) — also when the user says a page looks generic, templated, \"AI-made\", sloppy, or asks to make it look designed, premium, award-level or less like every other site. Not for dashboards or app screens.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7EWY8FQFNQ2GB7NZZWGSSDX","test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-tdd\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7EWY5SHBFRH1HDHVWM9K7TV","systematic-debugging","Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-debugging\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7EWY3NM0Z0JAKS7HDZ24K66","requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-code-review\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7EWY0WRMWX4M9ZFY0WKC593","brainstorming","You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-brainstorming\u002FSKILL.md",[],[]]