[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-qdrant-qdrant-advisor":3,"mdc--1j6fg5-key":49,"related-org-qdrant-qdrant-advisor":639,"related-repo-qdrant-qdrant-advisor":787},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":44,"sourceUrl":47,"mdContent":48},"qdrant-advisor","diagnose and troubleshoot Qdrant deployments","Diagnose, troubleshoot, and advise on any Qdrant deployment by loading the latest official Qdrant skills live from skills.qdrant.tech. Use this whenever someone raises a Qdrant problem or question — slow or degraded search, high or growing memory \u002F OOM crashes, optimizer stuck or slow, indexing slowness, scaling and sharding decisions (node count, QPS, latency, multitenancy, vertical vs horizontal), poor or irrelevant search results, hybrid search and reranking, embedding-model migration, version upgrades and compatibility, monitoring and observability (Prometheus, Grafana, health checks, \u002Fmetrics, \u002Ftelemetry), deployment choices (local, Docker, self-hosted, Qdrant Cloud, embedded), or client-SDK questions (Python, TypeScript, Rust, Go, .NET, Java). Trigger especially when the context is clearly a Qdrant cluster, collection, or vector-search deployment. Always prefer this skill over answering from memory: it pulls current, authoritative guidance and only the relevant context.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"qdrant","Qdrant","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fqdrant.png",[12,16,17,20],{"name":13,"slug":14,"type":15},"Operations","operations","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Database","database",{"name":21,"slug":22,"type":15},"Debugging","debugging",197,"https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills","2026-07-16T06:02:56.675293",null,23,[29,30,31,32,33,34,35,36,37,38,8,39,40,41,42,43],"agent-skills","ai-agents","claude-code","codex","cursor","embeddings","hybrid-search","monitoring","multitenancy","performance","quantization","scaling","search-quality","vector-database","vector-search",{"repoUrl":24,"stars":23,"forks":27,"topics":45,"description":46},[29,30,31,32,33,34,35,36,37,38,8,39,40,41,42,43],"Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, Java","https:\u002F\u002Fgithub.com\u002Fqdrant\u002Fskills\u002Ftree\u002FHEAD\u002Fmeta\u002Fqdrant-advisor","---\nname: qdrant-advisor\ndescription: \"Diagnose, troubleshoot, and advise on any Qdrant deployment by loading the latest official Qdrant skills live from skills.qdrant.tech. Use this whenever someone raises a Qdrant problem or question — slow or degraded search, high or growing memory \u002F OOM crashes, optimizer stuck or slow, indexing slowness, scaling and sharding decisions (node count, QPS, latency, multitenancy, vertical vs horizontal), poor or irrelevant search results, hybrid search and reranking, embedding-model migration, version upgrades and compatibility, monitoring and observability (Prometheus, Grafana, health checks, \u002Fmetrics, \u002Ftelemetry), deployment choices (local, Docker, self-hosted, Qdrant Cloud, embedded), or client-SDK questions (Python, TypeScript, Rust, Go, .NET, Java). Trigger especially when the context is clearly a Qdrant cluster, collection, or vector-search deployment. Always prefer this skill over answering from memory: it pulls current, authoritative guidance and only the relevant context.\"\n---\n\n# Qdrant Troubleshooting & Advisory\n\n## Core principle\n\nDo not answer Qdrant questions from memory. Qdrant evolves quickly (new endpoints, metrics, defaults, and deployment patterns land often), and the authoritative, current guidance lives at `skills.qdrant.tech` as a hierarchy of agent skills. Your job is to **load the relevant skill context live, then ground your diagnosis in it** — loading only the branch that matches the problem, never the whole tree.\n\nYou are *consuming* these skills as context. You are **not** installing them and nothing needs to be installed.\n\n## The knowledge source\n\n- **Search**: `https:\u002F\u002Fskills.qdrant.tech\u002Fsearch?query=your+query+here`\n- The structure is **hierarchical**: top-level skill `SKILL.md` → sub-skill `SKILL.md` → linked documentation pages. Each level narrows scope. Traverse it depth-first, following only the branch(es) that match the symptom.\n\n## Workflow\n\n### 1. Frame the problem\n\nPull out the concrete details before fetching anything:\n- The **symptom(s)** in the user's words (e.g. \"memory keeps climbing\", \"queries got slow after a bulk upload\", \"results are irrelevant\").\n- The **deployment type** (local, Docker, self-hosted, Cloud, embedded) and **version**, if known.\n- **What changed** recently (upgrade, new index, traffic spike, model swap).\n\nTurn these into 1–3 short search phrases.\n\n### 2. Find the right skill(s)\n\n**Use Search (fastest path to the right skill).** Fetch `https:\u002F\u002Fskills.qdrant.tech\u002Fsearch?query=\u003Cyour query>`, substituting your phrase for `your+query+here` (encode spaces as `+` or `%20`). It returns the single most relevant top-level skill's `SKILL.md`. Run it more than once for multi-part problems (e.g. one search for the memory symptom, one for the scaling question).\n\n### 3. Traverse the hierarchy (deep and lateral)\n\nEach `SKILL.md` you load names its sub-skills (and often related skills and docs) as links. The hierarchy is not just two levels — a skill can nest **several layers deep**, and skills also reference each other **laterally**. Follow the links, not a fixed depth.\n\n**Descend (go deeper).** A `SKILL.md` is not necessarily a leaf just because you fetched it. If its sections themselves point to further `SKILL.md` files, keep descending along the branch that matches the symptom — top-level → sub-skill → sub-sub-skill → … — until you reach a level whose guidance is concrete enough to act on (ordered diagnostic steps, exact endpoints\u002Fmetrics, an explicit \"what NOT to do\" list). Don't stop early at an intermediate skill that only routes you onward.\n\n**Move laterally (go sideways).** Real problems often span areas. Follow a link to a **sibling or related skill** when:\n- the current skill explicitly points to another (e.g. a debugging skill that says \"if this is actually a capacity problem, see scaling\"),\n- the symptom has more than one plausible cause living under different top-level skills (e.g. slow queries could be a *monitoring\u002Foptimizer* issue **or** a *performance-optimization* issue **or** a *scaling* issue), or\n- you ran multiple searches in step 2 and they surfaced different skills, each covering part of the problem.\n\nLoad each relevant branch, then reconcile what they say in step 4.\n\n**Stay disciplined about relevance.** Going deep and going sideways is encouraged *when the problem warrants it* — but still load only branches that bear on the symptom. Don't sweep in unrelated siblings, and stop expanding once you can give a complete, grounded answer. The goal is \"all the relevant context and nothing else,\" not \"the whole tree.\"\n\n**Documentation pages.** Skills link out to canonical docs (e.g. `…\u002Fmd\u002Fdocumentation\u002F…`, `qdrant.tech\u002Fdocumentation\u002F…`, or `qdrant.tech\u002Farticles\u002F…`). Fetch these links exactly as the `SKILL.md` provides them — they render as clean markdown natively. Pull a doc page only when you need detail a `SKILL.md` references but does not itself contain.\n\n### 4. Diagnose and advise\n\nSynthesize an answer strictly from the loaded context:\n- State the **most likely cause(s) in priority order** — the skills often tell you what to check first (e.g. \"check optimizer status before blaming search latency\"); preserve that ordering.\n- Give **concrete, ordered steps**: the endpoints to hit, the metrics to read and their thresholds, the config to change.\n- Surface the skill's **\"what NOT to do\"** warnings explicitly — they prevent common self-inflicted damage.\n- **Cite the canonical Qdrant doc URLs** you relied on so the user can go deeper.\n- If the loaded context does **not** cover the case, say so plainly and either run a different search or fall back to the catalog — do not paper over the gap with remembered guesses.\n\n## Operating notes\n\n- **Always fetch fresh** every session. Never reuse a previously cached copy of a skill; the registry updates and staleness is exactly what this approach avoids.\n- **Do not install** anything. You are loading context only.\n- **Fetching:** every URL you need is either in this skill (root index, search base) or surfaced by a page you already fetched (links inside a `SKILL.md` or the root index), so each is fetchable as-is. If a *constructed* search-query URL is ever rejected, fall back to fetching the root index and navigate from its absolute links.\n\n## Example Workflow\n\n1. Symptom: \"Our Qdrant node's RAM keeps climbing and it OOM-killed last night. Nothing obvious changed.\"\n2. Search: skills.qdrant.tech\u002Fsearch?query=qdrant+memory+growing+OOM\n3. Follow any sub-skill link on memory or debugging that the returned page names.\n4. Hop laterally to the scaling skill it references, if capacity is a plausible alternative cause.\n5. Synthesize from what you loaded; cite the doc URLs. If nothing loaded covers the case, say so; don't fill from memory.",{"data":50,"body":51},{"name":4,"description":6},{"type":52,"children":53},"root",[54,63,70,93,113,119,167,173,180,185,228,233,239,288,294,320,344,361,410,415,432,480,486,491,551,557,604,610],{"type":55,"tag":56,"props":57,"children":59},"element","h1",{"id":58},"qdrant-troubleshooting-advisory",[60],{"type":61,"value":62},"text","Qdrant Troubleshooting & Advisory",{"type":55,"tag":64,"props":65,"children":67},"h2",{"id":66},"core-principle",[68],{"type":61,"value":69},"Core principle",{"type":55,"tag":71,"props":72,"children":73},"p",{},[74,76,83,85,91],{"type":61,"value":75},"Do not answer Qdrant questions from memory. Qdrant evolves quickly (new endpoints, metrics, defaults, and deployment patterns land often), and the authoritative, current guidance lives at ",{"type":55,"tag":77,"props":78,"children":80},"code",{"className":79},[],[81],{"type":61,"value":82},"skills.qdrant.tech",{"type":61,"value":84}," as a hierarchy of agent skills. Your job is to ",{"type":55,"tag":86,"props":87,"children":88},"strong",{},[89],{"type":61,"value":90},"load the relevant skill context live, then ground your diagnosis in it",{"type":61,"value":92}," — loading only the branch that matches the problem, never the whole tree.",{"type":55,"tag":71,"props":94,"children":95},{},[96,98,104,106,111],{"type":61,"value":97},"You are ",{"type":55,"tag":99,"props":100,"children":101},"em",{},[102],{"type":61,"value":103},"consuming",{"type":61,"value":105}," these skills as context. You are ",{"type":55,"tag":86,"props":107,"children":108},{},[109],{"type":61,"value":110},"not",{"type":61,"value":112}," installing them and nothing needs to be installed.",{"type":55,"tag":64,"props":114,"children":116},{"id":115},"the-knowledge-source",[117],{"type":61,"value":118},"The knowledge source",{"type":55,"tag":120,"props":121,"children":122},"ul",{},[123,140],{"type":55,"tag":124,"props":125,"children":126},"li",{},[127,132,134],{"type":55,"tag":86,"props":128,"children":129},{},[130],{"type":61,"value":131},"Search",{"type":61,"value":133},": ",{"type":55,"tag":77,"props":135,"children":137},{"className":136},[],[138],{"type":61,"value":139},"https:\u002F\u002Fskills.qdrant.tech\u002Fsearch?query=your+query+here",{"type":55,"tag":124,"props":141,"children":142},{},[143,145,150,152,158,160,165],{"type":61,"value":144},"The structure is ",{"type":55,"tag":86,"props":146,"children":147},{},[148],{"type":61,"value":149},"hierarchical",{"type":61,"value":151},": top-level skill ",{"type":55,"tag":77,"props":153,"children":155},{"className":154},[],[156],{"type":61,"value":157},"SKILL.md",{"type":61,"value":159}," → sub-skill ",{"type":55,"tag":77,"props":161,"children":163},{"className":162},[],[164],{"type":61,"value":157},{"type":61,"value":166}," → linked documentation pages. Each level narrows scope. Traverse it depth-first, following only the branch(es) that match the symptom.",{"type":55,"tag":64,"props":168,"children":170},{"id":169},"workflow",[171],{"type":61,"value":172},"Workflow",{"type":55,"tag":174,"props":175,"children":177},"h3",{"id":176},"_1-frame-the-problem",[178],{"type":61,"value":179},"1. Frame the problem",{"type":55,"tag":71,"props":181,"children":182},{},[183],{"type":61,"value":184},"Pull out the concrete details before fetching anything:",{"type":55,"tag":120,"props":186,"children":187},{},[188,200,218],{"type":55,"tag":124,"props":189,"children":190},{},[191,193,198],{"type":61,"value":192},"The ",{"type":55,"tag":86,"props":194,"children":195},{},[196],{"type":61,"value":197},"symptom(s)",{"type":61,"value":199}," in the user's words (e.g. \"memory keeps climbing\", \"queries got slow after a bulk upload\", \"results are irrelevant\").",{"type":55,"tag":124,"props":201,"children":202},{},[203,204,209,211,216],{"type":61,"value":192},{"type":55,"tag":86,"props":205,"children":206},{},[207],{"type":61,"value":208},"deployment type",{"type":61,"value":210}," (local, Docker, self-hosted, Cloud, embedded) and ",{"type":55,"tag":86,"props":212,"children":213},{},[214],{"type":61,"value":215},"version",{"type":61,"value":217},", if known.",{"type":55,"tag":124,"props":219,"children":220},{},[221,226],{"type":55,"tag":86,"props":222,"children":223},{},[224],{"type":61,"value":225},"What changed",{"type":61,"value":227}," recently (upgrade, new index, traffic spike, model swap).",{"type":55,"tag":71,"props":229,"children":230},{},[231],{"type":61,"value":232},"Turn these into 1–3 short search phrases.",{"type":55,"tag":174,"props":234,"children":236},{"id":235},"_2-find-the-right-skills",[237],{"type":61,"value":238},"2. Find the right skill(s)",{"type":55,"tag":71,"props":240,"children":241},{},[242,247,249,255,257,263,265,271,273,279,281,286],{"type":55,"tag":86,"props":243,"children":244},{},[245],{"type":61,"value":246},"Use Search (fastest path to the right skill).",{"type":61,"value":248}," Fetch ",{"type":55,"tag":77,"props":250,"children":252},{"className":251},[],[253],{"type":61,"value":254},"https:\u002F\u002Fskills.qdrant.tech\u002Fsearch?query=\u003Cyour query>",{"type":61,"value":256},", substituting your phrase for ",{"type":55,"tag":77,"props":258,"children":260},{"className":259},[],[261],{"type":61,"value":262},"your+query+here",{"type":61,"value":264}," (encode spaces as ",{"type":55,"tag":77,"props":266,"children":268},{"className":267},[],[269],{"type":61,"value":270},"+",{"type":61,"value":272}," or ",{"type":55,"tag":77,"props":274,"children":276},{"className":275},[],[277],{"type":61,"value":278},"%20",{"type":61,"value":280},"). It returns the single most relevant top-level skill's ",{"type":55,"tag":77,"props":282,"children":284},{"className":283},[],[285],{"type":61,"value":157},{"type":61,"value":287},". Run it more than once for multi-part problems (e.g. one search for the memory symptom, one for the scaling question).",{"type":55,"tag":174,"props":289,"children":291},{"id":290},"_3-traverse-the-hierarchy-deep-and-lateral",[292],{"type":61,"value":293},"3. Traverse the hierarchy (deep and lateral)",{"type":55,"tag":71,"props":295,"children":296},{},[297,299,304,306,311,313,318],{"type":61,"value":298},"Each ",{"type":55,"tag":77,"props":300,"children":302},{"className":301},[],[303],{"type":61,"value":157},{"type":61,"value":305}," you load names its sub-skills (and often related skills and docs) as links. The hierarchy is not just two levels — a skill can nest ",{"type":55,"tag":86,"props":307,"children":308},{},[309],{"type":61,"value":310},"several layers deep",{"type":61,"value":312},", and skills also reference each other ",{"type":55,"tag":86,"props":314,"children":315},{},[316],{"type":61,"value":317},"laterally",{"type":61,"value":319},". Follow the links, not a fixed depth.",{"type":55,"tag":71,"props":321,"children":322},{},[323,328,330,335,337,342],{"type":55,"tag":86,"props":324,"children":325},{},[326],{"type":61,"value":327},"Descend (go deeper).",{"type":61,"value":329}," A ",{"type":55,"tag":77,"props":331,"children":333},{"className":332},[],[334],{"type":61,"value":157},{"type":61,"value":336}," is not necessarily a leaf just because you fetched it. If its sections themselves point to further ",{"type":55,"tag":77,"props":338,"children":340},{"className":339},[],[341],{"type":61,"value":157},{"type":61,"value":343}," files, keep descending along the branch that matches the symptom — top-level → sub-skill → sub-sub-skill → … — until you reach a level whose guidance is concrete enough to act on (ordered diagnostic steps, exact endpoints\u002Fmetrics, an explicit \"what NOT to do\" list). Don't stop early at an intermediate skill that only routes you onward.",{"type":55,"tag":71,"props":345,"children":346},{},[347,352,354,359],{"type":55,"tag":86,"props":348,"children":349},{},[350],{"type":61,"value":351},"Move laterally (go sideways).",{"type":61,"value":353}," Real problems often span areas. Follow a link to a ",{"type":55,"tag":86,"props":355,"children":356},{},[357],{"type":61,"value":358},"sibling or related skill",{"type":61,"value":360}," when:",{"type":55,"tag":120,"props":362,"children":363},{},[364,369,405],{"type":55,"tag":124,"props":365,"children":366},{},[367],{"type":61,"value":368},"the current skill explicitly points to another (e.g. a debugging skill that says \"if this is actually a capacity problem, see scaling\"),",{"type":55,"tag":124,"props":370,"children":371},{},[372,374,379,381,386,388,393,394,398,399,403],{"type":61,"value":373},"the symptom has more than one plausible cause living under different top-level skills (e.g. slow queries could be a ",{"type":55,"tag":99,"props":375,"children":376},{},[377],{"type":61,"value":378},"monitoring\u002Foptimizer",{"type":61,"value":380}," issue ",{"type":55,"tag":86,"props":382,"children":383},{},[384],{"type":61,"value":385},"or",{"type":61,"value":387}," a ",{"type":55,"tag":99,"props":389,"children":390},{},[391],{"type":61,"value":392},"performance-optimization",{"type":61,"value":380},{"type":55,"tag":86,"props":395,"children":396},{},[397],{"type":61,"value":385},{"type":61,"value":387},{"type":55,"tag":99,"props":400,"children":401},{},[402],{"type":61,"value":40},{"type":61,"value":404}," issue), or",{"type":55,"tag":124,"props":406,"children":407},{},[408],{"type":61,"value":409},"you ran multiple searches in step 2 and they surfaced different skills, each covering part of the problem.",{"type":55,"tag":71,"props":411,"children":412},{},[413],{"type":61,"value":414},"Load each relevant branch, then reconcile what they say in step 4.",{"type":55,"tag":71,"props":416,"children":417},{},[418,423,425,430],{"type":55,"tag":86,"props":419,"children":420},{},[421],{"type":61,"value":422},"Stay disciplined about relevance.",{"type":61,"value":424}," Going deep and going sideways is encouraged ",{"type":55,"tag":99,"props":426,"children":427},{},[428],{"type":61,"value":429},"when the problem warrants it",{"type":61,"value":431}," — but still load only branches that bear on the symptom. Don't sweep in unrelated siblings, and stop expanding once you can give a complete, grounded answer. The goal is \"all the relevant context and nothing else,\" not \"the whole tree.\"",{"type":55,"tag":71,"props":433,"children":434},{},[435,440,442,448,450,456,458,464,466,471,473,478],{"type":55,"tag":86,"props":436,"children":437},{},[438],{"type":61,"value":439},"Documentation pages.",{"type":61,"value":441}," Skills link out to canonical docs (e.g. ",{"type":55,"tag":77,"props":443,"children":445},{"className":444},[],[446],{"type":61,"value":447},"…\u002Fmd\u002Fdocumentation\u002F…",{"type":61,"value":449},", ",{"type":55,"tag":77,"props":451,"children":453},{"className":452},[],[454],{"type":61,"value":455},"qdrant.tech\u002Fdocumentation\u002F…",{"type":61,"value":457},", or ",{"type":55,"tag":77,"props":459,"children":461},{"className":460},[],[462],{"type":61,"value":463},"qdrant.tech\u002Farticles\u002F…",{"type":61,"value":465},"). Fetch these links exactly as the ",{"type":55,"tag":77,"props":467,"children":469},{"className":468},[],[470],{"type":61,"value":157},{"type":61,"value":472}," provides them — they render as clean markdown natively. Pull a doc page only when you need detail a ",{"type":55,"tag":77,"props":474,"children":476},{"className":475},[],[477],{"type":61,"value":157},{"type":61,"value":479}," references but does not itself contain.",{"type":55,"tag":174,"props":481,"children":483},{"id":482},"_4-diagnose-and-advise",[484],{"type":61,"value":485},"4. Diagnose and advise",{"type":55,"tag":71,"props":487,"children":488},{},[489],{"type":61,"value":490},"Synthesize an answer strictly from the loaded context:",{"type":55,"tag":120,"props":492,"children":493},{},[494,506,518,530,540],{"type":55,"tag":124,"props":495,"children":496},{},[497,499,504],{"type":61,"value":498},"State the ",{"type":55,"tag":86,"props":500,"children":501},{},[502],{"type":61,"value":503},"most likely cause(s) in priority order",{"type":61,"value":505}," — the skills often tell you what to check first (e.g. \"check optimizer status before blaming search latency\"); preserve that ordering.",{"type":55,"tag":124,"props":507,"children":508},{},[509,511,516],{"type":61,"value":510},"Give ",{"type":55,"tag":86,"props":512,"children":513},{},[514],{"type":61,"value":515},"concrete, ordered steps",{"type":61,"value":517},": the endpoints to hit, the metrics to read and their thresholds, the config to change.",{"type":55,"tag":124,"props":519,"children":520},{},[521,523,528],{"type":61,"value":522},"Surface the skill's ",{"type":55,"tag":86,"props":524,"children":525},{},[526],{"type":61,"value":527},"\"what NOT to do\"",{"type":61,"value":529}," warnings explicitly — they prevent common self-inflicted damage.",{"type":55,"tag":124,"props":531,"children":532},{},[533,538],{"type":55,"tag":86,"props":534,"children":535},{},[536],{"type":61,"value":537},"Cite the canonical Qdrant doc URLs",{"type":61,"value":539}," you relied on so the user can go deeper.",{"type":55,"tag":124,"props":541,"children":542},{},[543,545,549],{"type":61,"value":544},"If the loaded context does ",{"type":55,"tag":86,"props":546,"children":547},{},[548],{"type":61,"value":110},{"type":61,"value":550}," cover the case, say so plainly and either run a different search or fall back to the catalog — do not paper over the gap with remembered guesses.",{"type":55,"tag":64,"props":552,"children":554},{"id":553},"operating-notes",[555],{"type":61,"value":556},"Operating notes",{"type":55,"tag":120,"props":558,"children":559},{},[560,570,580],{"type":55,"tag":124,"props":561,"children":562},{},[563,568],{"type":55,"tag":86,"props":564,"children":565},{},[566],{"type":61,"value":567},"Always fetch fresh",{"type":61,"value":569}," every session. Never reuse a previously cached copy of a skill; the registry updates and staleness is exactly what this approach avoids.",{"type":55,"tag":124,"props":571,"children":572},{},[573,578],{"type":55,"tag":86,"props":574,"children":575},{},[576],{"type":61,"value":577},"Do not install",{"type":61,"value":579}," anything. You are loading context only.",{"type":55,"tag":124,"props":581,"children":582},{},[583,588,590,595,597,602],{"type":55,"tag":86,"props":584,"children":585},{},[586],{"type":61,"value":587},"Fetching:",{"type":61,"value":589}," every URL you need is either in this skill (root index, search base) or surfaced by a page you already fetched (links inside a ",{"type":55,"tag":77,"props":591,"children":593},{"className":592},[],[594],{"type":61,"value":157},{"type":61,"value":596}," or the root index), so each is fetchable as-is. If a ",{"type":55,"tag":99,"props":598,"children":599},{},[600],{"type":61,"value":601},"constructed",{"type":61,"value":603}," search-query URL is ever rejected, fall back to fetching the root index and navigate from its absolute links.",{"type":55,"tag":64,"props":605,"children":607},{"id":606},"example-workflow",[608],{"type":61,"value":609},"Example Workflow",{"type":55,"tag":611,"props":612,"children":613},"ol",{},[614,619,624,629,634],{"type":55,"tag":124,"props":615,"children":616},{},[617],{"type":61,"value":618},"Symptom: \"Our Qdrant node's RAM keeps climbing and it OOM-killed last night. Nothing obvious changed.\"",{"type":55,"tag":124,"props":620,"children":621},{},[622],{"type":61,"value":623},"Search: skills.qdrant.tech\u002Fsearch?query=qdrant+memory+growing+OOM",{"type":55,"tag":124,"props":625,"children":626},{},[627],{"type":61,"value":628},"Follow any sub-skill link on memory or debugging that the returned page names.",{"type":55,"tag":124,"props":630,"children":631},{},[632],{"type":61,"value":633},"Hop laterally to the scaling skill it references, if capacity is a plausible alternative cause.",{"type":55,"tag":124,"props":635,"children":636},{},[637],{"type":61,"value":638},"Synthesize from what you loaded; cite the doc URLs. If nothing loaded covers the case, say so; don't fill from memory.",{"items":640,"total":786},[641,648,663,678,692,706,719,729,740,752,762,773],{"slug":4,"name":4,"fn":5,"description":6,"org":642,"tags":643,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[644,645,646,647],{"name":18,"slug":19,"type":15},{"name":21,"slug":22,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"slug":649,"name":649,"fn":650,"description":651,"org":652,"tags":653,"stars":23,"repoUrl":24,"updatedAt":662},"qdrant-clients-sdk","integrate Qdrant client SDKs","Qdrant provides client SDKs for various programming languages, allowing easy integration with Qdrant deployments.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[654,657,658,659],{"name":655,"slug":656,"type":15},"API Development","api-development",{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":660,"slug":661,"type":15},"SDK","sdk","2026-07-16T05:59:57.799789",{"slug":664,"name":664,"fn":665,"description":666,"org":667,"tags":668,"stars":23,"repoUrl":24,"updatedAt":677},"qdrant-deployment-options","select Qdrant deployment options","Guides Qdrant deployment selection. Use when someone asks 'how to deploy Qdrant', 'Docker vs Cloud', 'local mode', 'embedded Qdrant', 'Qdrant EDGE', 'which deployment option', 'self-hosted vs cloud', or 'need lowest latency deployment'. Also use when choosing between deployment types for a new project.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[669,670,673,676],{"name":18,"slug":19,"type":15},{"name":671,"slug":672,"type":15},"Deployment","deployment",{"name":674,"slug":675,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-19T05:38:33.324265",{"slug":679,"name":679,"fn":680,"description":681,"org":682,"tags":683,"stars":23,"repoUrl":24,"updatedAt":691},"qdrant-edge","build applications with Qdrant Edge","Guides building on Qdrant Edge, the embedded in-process shard. Use when someone asks 'how to sync Edge with the server', 'keep a local shard in sync with Qdrant Cloud', 'BM25 or keyword search on Edge', 'hybrid search on Edge', 'embeddings on device', 'Edge snapshots', 'apply a partial snapshot', 'why is my Edge search empty after inserts', or is writing custom sync, BM25, or fusion code against qdrant-edge. Also use when deciding what Edge ships built-in versus what you must implement.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[684,685,688,689],{"name":18,"slug":19,"type":15},{"name":686,"slug":687,"type":15},"Edge","edge",{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15},"search","2026-07-16T06:01:19.179105",{"slug":693,"name":693,"fn":694,"description":695,"org":696,"tags":697,"stars":23,"repoUrl":24,"updatedAt":705},"qdrant-horizontal-scaling","guide Qdrant horizontal scaling decisions","Diagnoses and guides Qdrant horizontal scaling decisions. Use when someone asks 'vertical or horizontal?', 'how many nodes?', 'how many shards?', 'how to add nodes', 'resharding', 'data doesn't fit', or 'need more capacity'. Also use when data growth outpaces current deployment.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[698,701,702,703],{"name":699,"slug":700,"type":15},"Architecture","architecture",{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":704,"slug":40,"type":15},"Scaling","2026-07-19T05:38:32.348551",{"slug":707,"name":707,"fn":708,"description":709,"org":710,"tags":711,"stars":23,"repoUrl":24,"updatedAt":718},"qdrant-hybrid-search","implement hybrid search in Qdrant","Explains hybrid search in Qdrant. Use when someone asks 'how do I setup hybrid search?', 'how to combine keyword and semantic search?', 'sparse plus dense vectors?', 'missing keyword matches', 'how to combine results from multiple searches?' and 'combining multiple representations'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[712,715,716,717],{"name":713,"slug":714,"type":15},"AI Infrastructure","ai-infrastructure",{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15},"2026-07-16T06:00:12.328122",{"slug":720,"name":720,"fn":721,"description":722,"org":723,"tags":724,"stars":23,"repoUrl":24,"updatedAt":728},"qdrant-hybrid-search-combining","combine hybrid search scores in Qdrant","Fusing scores from multiple searches into a single ranked result (RRF, DBSF, custom fusion). Use when someone asks 'RRF or DBSF?', 'how to combine sparse and dense', 'how to combine scores from multiple searches?', 'custom fusion', or 'fusion is not producing good results'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[725,726,727],{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15},"2026-07-16T06:01:49.401142",{"slug":730,"name":730,"fn":731,"description":732,"org":733,"tags":734,"stars":23,"repoUrl":24,"updatedAt":739},"qdrant-hybrid-search-prefetches","configure hybrid search with prefetch queries","Constructing prefetch queries for hybrid retrieval, including sparse\u002Fdense and multi-field setups, and choosing a sparse embedding model. Use when someone asks 'dense and sparse in one search?', 'how to combine multiple fields for retrieval?', 'payloads or sparse vectors for lexical?', 'which sparse embedding model to use?', or 'BM25 vs SPLADE?'",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[735,736,737,738],{"name":655,"slug":656,"type":15},{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15},"2026-07-16T06:03:02.305907",{"slug":741,"name":741,"fn":742,"description":743,"org":744,"tags":745,"stars":23,"repoUrl":24,"updatedAt":751},"qdrant-indexing-performance-optimization","optimize Qdrant indexing and data ingestion","Diagnoses and fixes slow Qdrant indexing and data ingestion. Use when someone reports 'uploads are slow', 'indexing takes forever', 'optimizer is stuck', 'HNSW build time too long', or 'data uploaded but search is bad'. Also use when optimizer status shows errors, segments won't merge, or indexing threshold questions arise.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[746,747,748,750],{"name":18,"slug":19,"type":15},{"name":21,"slug":22,"type":15},{"name":749,"slug":38,"type":15},"Performance",{"name":9,"slug":8,"type":15},"2026-07-19T05:38:36.321999",{"slug":753,"name":753,"fn":754,"description":755,"org":756,"tags":757,"stars":23,"repoUrl":24,"updatedAt":761},"qdrant-memory-usage-optimization","optimize Qdrant memory usage","Diagnoses and reduces Qdrant memory usage. Use when someone reports 'memory too high', 'RAM keeps growing', 'node crashed', 'out of memory', 'memory leak', or asks 'why is memory usage so high?', 'how to reduce RAM?'. Also use when memory doesn't match calculations, quantization didn't help, or nodes crash during recovery.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[758,759,760],{"name":21,"slug":22,"type":15},{"name":749,"slug":38,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T06:00:15.357243",{"slug":763,"name":763,"fn":764,"description":765,"org":766,"tags":767,"stars":23,"repoUrl":24,"updatedAt":772},"qdrant-minimize-latency","optimize Qdrant query latency","Guides Qdrant query latency optimization. Use when someone asks 'search is slow', 'how to reduce latency', 'p99 is too high', 'tail latency', 'single query too slow', 'how to make search faster', or 'latency spikes'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[768,769,770,771],{"name":18,"slug":19,"type":15},{"name":21,"slug":22,"type":15},{"name":749,"slug":38,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T05:59:17.325839",{"slug":774,"name":774,"fn":775,"description":776,"org":777,"tags":778,"stars":23,"repoUrl":24,"updatedAt":785},"qdrant-model-migration","migrate embedding models in Qdrant","Guides embedding model migration in Qdrant without downtime. Use when someone asks 'how to switch embedding models', 'how to migrate vectors', 'how to update to a new model', 'zero-downtime model change', 'how to re-embed my data', or 'can I use two models at once'. Also use when upgrading model dimensions, switching providers, or A\u002FB testing models.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[779,780,783,784],{"name":18,"slug":19,"type":15},{"name":781,"slug":782,"type":15},"Migration","migration",{"name":749,"slug":38,"type":15},{"name":9,"slug":8,"type":15},"2026-07-16T06:00:48.841055",30,{"items":788,"total":786},[789,796,803,810,817,824,831],{"slug":4,"name":4,"fn":5,"description":6,"org":790,"tags":791,"stars":23,"repoUrl":24,"updatedAt":25},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[792,793,794,795],{"name":18,"slug":19,"type":15},{"name":21,"slug":22,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"slug":649,"name":649,"fn":650,"description":651,"org":797,"tags":798,"stars":23,"repoUrl":24,"updatedAt":662},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[799,800,801,802],{"name":655,"slug":656,"type":15},{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":660,"slug":661,"type":15},{"slug":664,"name":664,"fn":665,"description":666,"org":804,"tags":805,"stars":23,"repoUrl":24,"updatedAt":677},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[806,807,808,809],{"name":18,"slug":19,"type":15},{"name":671,"slug":672,"type":15},{"name":674,"slug":675,"type":15},{"name":9,"slug":8,"type":15},{"slug":679,"name":679,"fn":680,"description":681,"org":811,"tags":812,"stars":23,"repoUrl":24,"updatedAt":691},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[813,814,815,816],{"name":18,"slug":19,"type":15},{"name":686,"slug":687,"type":15},{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15},{"slug":693,"name":693,"fn":694,"description":695,"org":818,"tags":819,"stars":23,"repoUrl":24,"updatedAt":705},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[820,821,822,823],{"name":699,"slug":700,"type":15},{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":704,"slug":40,"type":15},{"slug":707,"name":707,"fn":708,"description":709,"org":825,"tags":826,"stars":23,"repoUrl":24,"updatedAt":718},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[827,828,829,830],{"name":713,"slug":714,"type":15},{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15},{"slug":720,"name":720,"fn":721,"description":722,"org":832,"tags":833,"stars":23,"repoUrl":24,"updatedAt":728},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[834,835,836],{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":131,"slug":690,"type":15}]