[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-clickhouse-clickhouse-js-node-rowbinary":3,"mdc-piz2rs-key":38,"related-org-clickhouse-clickhouse-js-node-rowbinary":696,"related-repo-clickhouse-clickhouse-js-node-rowbinary":889},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":33,"sourceUrl":36,"mdContent":37},"clickhouse-js-node-rowbinary","encode and decode ClickHouse RowBinary streams","Generate TypeScript\u002FJavaScript code that reads\u002Fdecodes AND writes\u002Fencodes ClickHouse RowBinary streams for the ClickHouse HTTP server. Use this skill whenever a user wants to parse or produce `RowBinary`, `RowBinaryWithNames`, or `RowBinaryWithNamesAndTypes`. Node.js only, doesn't cover browsers.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"clickhouse","ClickHouse","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fclickhouse.png",[12,14,17,20,23],{"name":9,"slug":8,"type":13},"tag",{"name":15,"slug":16,"type":13},"TypeScript","typescript",{"name":18,"slug":19,"type":13},"JavaScript","javascript",{"name":21,"slug":22,"type":13},"Database","database",{"name":24,"slug":25,"type":13},"API Development","api-development",493,"https:\u002F\u002Fgithub.com\u002FClickHouse\u002Fagent-skills","2026-07-01T08:16:19.314713",null,32,[32,8],"agents",{"repoUrl":27,"stars":26,"forks":30,"topics":34,"description":35},[32,8],"The official Agent Skills for ClickHouse and ClickHouse Cloud","https:\u002F\u002Fgithub.com\u002FClickHouse\u002Fagent-skills\u002Ftree\u002FHEAD\u002Fskills\u002Fclickhouse-js-node-rowbinary","---\nname: clickhouse-js-node-rowbinary\ndescription: >\n  Generate TypeScript\u002FJavaScript code that reads\u002Fdecodes AND writes\u002Fencodes\n  ClickHouse RowBinary streams for the ClickHouse HTTP server.\n  Use this skill whenever a user wants to parse or produce `RowBinary`,\n  `RowBinaryWithNames`, or `RowBinaryWithNamesAndTypes`.\n  Node.js only, doesn't cover browsers.\n---\n\n# ClickHouse JS RowBinary Codec Generator for Node.js\n\nThis skill generates both directions of the wire format: **readers** (decode\nbytes → values) and **writers** (encode values → bytes, the mirror). A given\ntask normally needs only one side. This file is the shared entry point — the\nformat gate plus the principles common to both directions; the per-direction\ndecisions, guidance, and the per-type reference tables live in two sibling files.\n\n**Pick your side — read only the one you need:**\n\n- **Decoding a `RowBinary*` response** from ClickHouse into JS values →\n  **[reader.md](reader.md)**. Streaming vs whole-buffer, row-objects vs columnar,\n  fixed vs runtime schema, and the per-type reader reference.\n- **Encoding JS values into a `RowBinary` payload** to send to ClickHouse →\n  **[writer.md](writer.md)**. The `Sink`\u002F`writeX` building blocks, `writeRows`\n  streaming, and the per-type writer reference.\n\nThe per-type code is real, split by direction under `src\u002Freaders\u002F` and\n`src\u002Fwriters\u002F`.\n\n## First: is RowBinary even the right format?\n\nRowBinary exists for throughput, but it is **not automatically the fastest\npath** — match the format to the shape of the data before committing to a\nbespoke parser.\n\n**Prefer a `JSON*` format (e.g. `JSONEachRow`) when** the result is mostly\nstrings \u002F JSON-like values that you consume wholesale — randomly accessing\nessentially every field, running string\u002Fregexp methods on them, treating values\nas text. V8's native `JSON.parse` is heavily optimized C++ and builds JS strings\nand objects faster than a JS-level RowBinary decoder can; pair it with HTTP\nresponse compression (`gzip` \u002F `zstd`, which crushes JSON's repetitive keys) and\nthe wire cost shrinks too.\n\n**RowBinary clearly wins when** the result is dominated by:\n\n- **Wide numerics** — `Int128`\u002F`Int256`\u002F`UInt128`\u002F`UInt256`,\n  `Decimal128`\u002F`Decimal256`.\n- **Binary \u002F fixed-width blobs** — `IPv4`, `IPv6`, `UUID`, `FixedString`.\n- **High-volume fixed-width numeric columns** generally, where each value is a\n  single `DataView` read.\n\n**Prefer the `Native` format when** columnar load and client-side analytics are\nthe main goal (fold\u002Fscan\u002Ffilter columns, feed typed arrays to a Worker or WASM).\n`Native` is column-major, so it loads straight into one typed array per column\nwith no transpose.\n\nFor help choosing and consuming a `JSON*` format (or CSV \u002F TSV) instead, use the\n**`clickhouse-js-node-coding`** skill.\n\n## Core guidance (both directions)\n\nThese principles apply whether you are generating a reader or a writer; the\nside-specific operational guidance is in [reader.md](reader.md) \u002F\n[writer.md](writer.md).\n\n- **Little-endian only.** RowBinary is little-endian; target x86\u002FARM. Read and\n  write every multi-byte number with `DataView` accessors passing a **literal**\n  `true` for the `littleEndian` flag.\n\n- **Correct first, then optimize.** First emit a correct codec built from the\n  plain per-type API. Only after it's correct (and tested) specialize it. Don't\n  bake performance assumptions in before correctness.\n\n- **Monomorphize generic\u002Fcomposite types.** Emit specialized, inlined code per\n  type combination instead of passing functions as arguments where the type is\n  known ahead of time.\n\n- **Inline the leaf ops.** The per-type `readX`\u002F`writeX` functions are the\n  correct, composable reference; the generated codec should INLINE their bodies,\n  not call them, so the row loop is straight-line with no per-field indirection\n  (and so the fixed-width coalescing can fold the offset arithmetic together).\n\n- **Annotate the type per column.** Inlining erases the type structure, so put a\n  short comment above each column's encode\u002Fdecode block naming the ClickHouse\n  type it handles.\n\n- **Shared scratch is not reentrant.** Some hot methods reuse a module-level\n  scratch buffer as a write-then-read pair — correct only because the access is\n  fully synchronous. An `async`\u002F`yield` boundary between populating and reading\n  it corrupts the value.\n\n- **TypeScript by default.** Generate TypeScript code and helpers unless the user\n  explicitly asks for plain JavaScript.\n\n## Worked examples\n\nSix end-to-end examples with real speedup are catalogued in [EXAMPLES.md](EXAMPLES.md).\n\n## Out of scope\n\n- **JSON \u002F CSV \u002F TSV \u002F Parquet parsing** → use `clickhouse-js-node-coding`.\n- **Connection errors, hangs, type mismatches** → use\n  `clickhouse-js-node-troubleshooting`.\n- **Browser \u002F Web Worker \u002F Edge** → `@clickhouse\u002Fclient-web`.\n\n## Still Stuck?\n\n- [ClickHouse RowBinary format](https:\u002F\u002Fclickhouse.com\u002Fdocs\u002Finterfaces\u002Fformats#rowbinary)\n- [ClickHouse data types](https:\u002F\u002Fclickhouse.com\u002Fdocs\u002Fsql-reference\u002Fdata-types)\n- [ClickHouse JS client docs](https:\u002F\u002Fclickhouse.com\u002Fdocs\u002Fintegrations\u002Fjavascript)\n",{"data":39,"body":40},{"name":4,"description":6},{"type":41,"children":42},"root",[43,52,73,81,168,189,196,208,258,268,380,405,428,434,450,580,586,597,603,656,662],{"type":44,"tag":45,"props":46,"children":48},"element","h1",{"id":47},"clickhouse-js-rowbinary-codec-generator-for-nodejs",[49],{"type":50,"value":51},"text","ClickHouse JS RowBinary Codec Generator for Node.js",{"type":44,"tag":53,"props":54,"children":55},"p",{},[56,58,64,66,71],{"type":50,"value":57},"This skill generates both directions of the wire format: ",{"type":44,"tag":59,"props":60,"children":61},"strong",{},[62],{"type":50,"value":63},"readers",{"type":50,"value":65}," (decode\nbytes → values) and ",{"type":44,"tag":59,"props":67,"children":68},{},[69],{"type":50,"value":70},"writers",{"type":50,"value":72}," (encode values → bytes, the mirror). A given\ntask normally needs only one side. This file is the shared entry point — the\nformat gate plus the principles common to both directions; the per-direction\ndecisions, guidance, and the per-type reference tables live in two sibling files.",{"type":44,"tag":53,"props":74,"children":75},{},[76],{"type":44,"tag":59,"props":77,"children":78},{},[79],{"type":50,"value":80},"Pick your side — read only the one you need:",{"type":44,"tag":82,"props":83,"children":84},"ul",{},[85,116],{"type":44,"tag":86,"props":87,"children":88},"li",{},[89,103,105,114],{"type":44,"tag":59,"props":90,"children":91},{},[92,94,101],{"type":50,"value":93},"Decoding a ",{"type":44,"tag":95,"props":96,"children":98},"code",{"className":97},[],[99],{"type":50,"value":100},"RowBinary*",{"type":50,"value":102}," response",{"type":50,"value":104}," from ClickHouse into JS values →\n",{"type":44,"tag":59,"props":106,"children":107},{},[108],{"type":44,"tag":109,"props":110,"children":112},"a",{"href":111},"reader.md",[113],{"type":50,"value":111},{"type":50,"value":115},". Streaming vs whole-buffer, row-objects vs columnar,\nfixed vs runtime schema, and the per-type reader reference.",{"type":44,"tag":86,"props":117,"children":118},{},[119,132,134,142,144,150,152,158,160,166],{"type":44,"tag":59,"props":120,"children":121},{},[122,124,130],{"type":50,"value":123},"Encoding JS values into a ",{"type":44,"tag":95,"props":125,"children":127},{"className":126},[],[128],{"type":50,"value":129},"RowBinary",{"type":50,"value":131}," payload",{"type":50,"value":133}," to send to ClickHouse →\n",{"type":44,"tag":59,"props":135,"children":136},{},[137],{"type":44,"tag":109,"props":138,"children":140},{"href":139},"writer.md",[141],{"type":50,"value":139},{"type":50,"value":143},". The ",{"type":44,"tag":95,"props":145,"children":147},{"className":146},[],[148],{"type":50,"value":149},"Sink",{"type":50,"value":151},"\u002F",{"type":44,"tag":95,"props":153,"children":155},{"className":154},[],[156],{"type":50,"value":157},"writeX",{"type":50,"value":159}," building blocks, ",{"type":44,"tag":95,"props":161,"children":163},{"className":162},[],[164],{"type":50,"value":165},"writeRows",{"type":50,"value":167},"\nstreaming, and the per-type writer reference.",{"type":44,"tag":53,"props":169,"children":170},{},[171,173,179,181,187],{"type":50,"value":172},"The per-type code is real, split by direction under ",{"type":44,"tag":95,"props":174,"children":176},{"className":175},[],[177],{"type":50,"value":178},"src\u002Freaders\u002F",{"type":50,"value":180}," and\n",{"type":44,"tag":95,"props":182,"children":184},{"className":183},[],[185],{"type":50,"value":186},"src\u002Fwriters\u002F",{"type":50,"value":188},".",{"type":44,"tag":190,"props":191,"children":193},"h2",{"id":192},"first-is-rowbinary-even-the-right-format",[194],{"type":50,"value":195},"First: is RowBinary even the right format?",{"type":44,"tag":53,"props":197,"children":198},{},[199,201,206],{"type":50,"value":200},"RowBinary exists for throughput, but it is ",{"type":44,"tag":59,"props":202,"children":203},{},[204],{"type":50,"value":205},"not automatically the fastest\npath",{"type":50,"value":207}," — match the format to the shape of the data before committing to a\nbespoke parser.",{"type":44,"tag":53,"props":209,"children":210},{},[211,232,234,240,242,248,250,256],{"type":44,"tag":59,"props":212,"children":213},{},[214,216,222,224,230],{"type":50,"value":215},"Prefer a ",{"type":44,"tag":95,"props":217,"children":219},{"className":218},[],[220],{"type":50,"value":221},"JSON*",{"type":50,"value":223}," format (e.g. ",{"type":44,"tag":95,"props":225,"children":227},{"className":226},[],[228],{"type":50,"value":229},"JSONEachRow",{"type":50,"value":231},") when",{"type":50,"value":233}," the result is mostly\nstrings \u002F JSON-like values that you consume wholesale — randomly accessing\nessentially every field, running string\u002Fregexp methods on them, treating values\nas text. V8's native ",{"type":44,"tag":95,"props":235,"children":237},{"className":236},[],[238],{"type":50,"value":239},"JSON.parse",{"type":50,"value":241}," is heavily optimized C++ and builds JS strings\nand objects faster than a JS-level RowBinary decoder can; pair it with HTTP\nresponse compression (",{"type":44,"tag":95,"props":243,"children":245},{"className":244},[],[246],{"type":50,"value":247},"gzip",{"type":50,"value":249}," \u002F ",{"type":44,"tag":95,"props":251,"children":253},{"className":252},[],[254],{"type":50,"value":255},"zstd",{"type":50,"value":257},", which crushes JSON's repetitive keys) and\nthe wire cost shrinks too.",{"type":44,"tag":53,"props":259,"children":260},{},[261,266],{"type":44,"tag":59,"props":262,"children":263},{},[264],{"type":50,"value":265},"RowBinary clearly wins when",{"type":50,"value":267}," the result is dominated by:",{"type":44,"tag":82,"props":269,"children":270},{},[271,324,362],{"type":44,"tag":86,"props":272,"children":273},{},[274,279,281,287,288,294,295,301,302,308,310,316,317,323],{"type":44,"tag":59,"props":275,"children":276},{},[277],{"type":50,"value":278},"Wide numerics",{"type":50,"value":280}," — ",{"type":44,"tag":95,"props":282,"children":284},{"className":283},[],[285],{"type":50,"value":286},"Int128",{"type":50,"value":151},{"type":44,"tag":95,"props":289,"children":291},{"className":290},[],[292],{"type":50,"value":293},"Int256",{"type":50,"value":151},{"type":44,"tag":95,"props":296,"children":298},{"className":297},[],[299],{"type":50,"value":300},"UInt128",{"type":50,"value":151},{"type":44,"tag":95,"props":303,"children":305},{"className":304},[],[306],{"type":50,"value":307},"UInt256",{"type":50,"value":309},",\n",{"type":44,"tag":95,"props":311,"children":313},{"className":312},[],[314],{"type":50,"value":315},"Decimal128",{"type":50,"value":151},{"type":44,"tag":95,"props":318,"children":320},{"className":319},[],[321],{"type":50,"value":322},"Decimal256",{"type":50,"value":188},{"type":44,"tag":86,"props":325,"children":326},{},[327,332,333,339,341,347,348,354,355,361],{"type":44,"tag":59,"props":328,"children":329},{},[330],{"type":50,"value":331},"Binary \u002F fixed-width blobs",{"type":50,"value":280},{"type":44,"tag":95,"props":334,"children":336},{"className":335},[],[337],{"type":50,"value":338},"IPv4",{"type":50,"value":340},", ",{"type":44,"tag":95,"props":342,"children":344},{"className":343},[],[345],{"type":50,"value":346},"IPv6",{"type":50,"value":340},{"type":44,"tag":95,"props":349,"children":351},{"className":350},[],[352],{"type":50,"value":353},"UUID",{"type":50,"value":340},{"type":44,"tag":95,"props":356,"children":358},{"className":357},[],[359],{"type":50,"value":360},"FixedString",{"type":50,"value":188},{"type":44,"tag":86,"props":363,"children":364},{},[365,370,372,378],{"type":44,"tag":59,"props":366,"children":367},{},[368],{"type":50,"value":369},"High-volume fixed-width numeric columns",{"type":50,"value":371}," generally, where each value is a\nsingle ",{"type":44,"tag":95,"props":373,"children":375},{"className":374},[],[376],{"type":50,"value":377},"DataView",{"type":50,"value":379}," read.",{"type":44,"tag":53,"props":381,"children":382},{},[383,396,398,403],{"type":44,"tag":59,"props":384,"children":385},{},[386,388,394],{"type":50,"value":387},"Prefer the ",{"type":44,"tag":95,"props":389,"children":391},{"className":390},[],[392],{"type":50,"value":393},"Native",{"type":50,"value":395}," format when",{"type":50,"value":397}," columnar load and client-side analytics are\nthe main goal (fold\u002Fscan\u002Ffilter columns, feed typed arrays to a Worker or WASM).\n",{"type":44,"tag":95,"props":399,"children":401},{"className":400},[],[402],{"type":50,"value":393},{"type":50,"value":404}," is column-major, so it loads straight into one typed array per column\nwith no transpose.",{"type":44,"tag":53,"props":406,"children":407},{},[408,410,415,417,426],{"type":50,"value":409},"For help choosing and consuming a ",{"type":44,"tag":95,"props":411,"children":413},{"className":412},[],[414],{"type":50,"value":221},{"type":50,"value":416}," format (or CSV \u002F TSV) instead, use the\n",{"type":44,"tag":59,"props":418,"children":419},{},[420],{"type":44,"tag":95,"props":421,"children":423},{"className":422},[],[424],{"type":50,"value":425},"clickhouse-js-node-coding",{"type":50,"value":427}," skill.",{"type":44,"tag":190,"props":429,"children":431},{"id":430},"core-guidance-both-directions",[432],{"type":50,"value":433},"Core guidance (both directions)",{"type":44,"tag":53,"props":435,"children":436},{},[437,439,443,445,449],{"type":50,"value":438},"These principles apply whether you are generating a reader or a writer; the\nside-specific operational guidance is in ",{"type":44,"tag":109,"props":440,"children":441},{"href":111},[442],{"type":50,"value":111},{"type":50,"value":444}," \u002F\n",{"type":44,"tag":109,"props":446,"children":447},{"href":139},[448],{"type":50,"value":139},{"type":50,"value":188},{"type":44,"tag":82,"props":451,"children":452},{},[453,491,501,511,535,545,570],{"type":44,"tag":86,"props":454,"children":455},{},[456,461,463,468,470,475,481,483,489],{"type":44,"tag":59,"props":457,"children":458},{},[459],{"type":50,"value":460},"Little-endian only.",{"type":50,"value":462}," RowBinary is little-endian; target x86\u002FARM. Read and\nwrite every multi-byte number with ",{"type":44,"tag":95,"props":464,"children":466},{"className":465},[],[467],{"type":50,"value":377},{"type":50,"value":469}," accessors passing a ",{"type":44,"tag":59,"props":471,"children":472},{},[473],{"type":50,"value":474},"literal",{"type":44,"tag":95,"props":476,"children":478},{"className":477},[],[479],{"type":50,"value":480},"true",{"type":50,"value":482}," for the ",{"type":44,"tag":95,"props":484,"children":486},{"className":485},[],[487],{"type":50,"value":488},"littleEndian",{"type":50,"value":490}," flag.",{"type":44,"tag":86,"props":492,"children":493},{},[494,499],{"type":44,"tag":59,"props":495,"children":496},{},[497],{"type":50,"value":498},"Correct first, then optimize.",{"type":50,"value":500}," First emit a correct codec built from the\nplain per-type API. Only after it's correct (and tested) specialize it. Don't\nbake performance assumptions in before correctness.",{"type":44,"tag":86,"props":502,"children":503},{},[504,509],{"type":44,"tag":59,"props":505,"children":506},{},[507],{"type":50,"value":508},"Monomorphize generic\u002Fcomposite types.",{"type":50,"value":510}," Emit specialized, inlined code per\ntype combination instead of passing functions as arguments where the type is\nknown ahead of time.",{"type":44,"tag":86,"props":512,"children":513},{},[514,519,521,527,528,533],{"type":44,"tag":59,"props":515,"children":516},{},[517],{"type":50,"value":518},"Inline the leaf ops.",{"type":50,"value":520}," The per-type ",{"type":44,"tag":95,"props":522,"children":524},{"className":523},[],[525],{"type":50,"value":526},"readX",{"type":50,"value":151},{"type":44,"tag":95,"props":529,"children":531},{"className":530},[],[532],{"type":50,"value":157},{"type":50,"value":534}," functions are the\ncorrect, composable reference; the generated codec should INLINE their bodies,\nnot call them, so the row loop is straight-line with no per-field indirection\n(and so the fixed-width coalescing can fold the offset arithmetic together).",{"type":44,"tag":86,"props":536,"children":537},{},[538,543],{"type":44,"tag":59,"props":539,"children":540},{},[541],{"type":50,"value":542},"Annotate the type per column.",{"type":50,"value":544}," Inlining erases the type structure, so put a\nshort comment above each column's encode\u002Fdecode block naming the ClickHouse\ntype it handles.",{"type":44,"tag":86,"props":546,"children":547},{},[548,553,555,561,562,568],{"type":44,"tag":59,"props":549,"children":550},{},[551],{"type":50,"value":552},"Shared scratch is not reentrant.",{"type":50,"value":554}," Some hot methods reuse a module-level\nscratch buffer as a write-then-read pair — correct only because the access is\nfully synchronous. An ",{"type":44,"tag":95,"props":556,"children":558},{"className":557},[],[559],{"type":50,"value":560},"async",{"type":50,"value":151},{"type":44,"tag":95,"props":563,"children":565},{"className":564},[],[566],{"type":50,"value":567},"yield",{"type":50,"value":569}," boundary between populating and reading\nit corrupts the value.",{"type":44,"tag":86,"props":571,"children":572},{},[573,578],{"type":44,"tag":59,"props":574,"children":575},{},[576],{"type":50,"value":577},"TypeScript by default.",{"type":50,"value":579}," Generate TypeScript code and helpers unless the user\nexplicitly asks for plain JavaScript.",{"type":44,"tag":190,"props":581,"children":583},{"id":582},"worked-examples",[584],{"type":50,"value":585},"Worked examples",{"type":44,"tag":53,"props":587,"children":588},{},[589,591,596],{"type":50,"value":590},"Six end-to-end examples with real speedup are catalogued in ",{"type":44,"tag":109,"props":592,"children":594},{"href":593},"EXAMPLES.md",[595],{"type":50,"value":593},{"type":50,"value":188},{"type":44,"tag":190,"props":598,"children":600},{"id":599},"out-of-scope",[601],{"type":50,"value":602},"Out of scope",{"type":44,"tag":82,"props":604,"children":605},{},[606,622,639],{"type":44,"tag":86,"props":607,"children":608},{},[609,614,616,621],{"type":44,"tag":59,"props":610,"children":611},{},[612],{"type":50,"value":613},"JSON \u002F CSV \u002F TSV \u002F Parquet parsing",{"type":50,"value":615}," → use ",{"type":44,"tag":95,"props":617,"children":619},{"className":618},[],[620],{"type":50,"value":425},{"type":50,"value":188},{"type":44,"tag":86,"props":623,"children":624},{},[625,630,632,638],{"type":44,"tag":59,"props":626,"children":627},{},[628],{"type":50,"value":629},"Connection errors, hangs, type mismatches",{"type":50,"value":631}," → use\n",{"type":44,"tag":95,"props":633,"children":635},{"className":634},[],[636],{"type":50,"value":637},"clickhouse-js-node-troubleshooting",{"type":50,"value":188},{"type":44,"tag":86,"props":640,"children":641},{},[642,647,649,655],{"type":44,"tag":59,"props":643,"children":644},{},[645],{"type":50,"value":646},"Browser \u002F Web Worker \u002F Edge",{"type":50,"value":648}," → ",{"type":44,"tag":95,"props":650,"children":652},{"className":651},[],[653],{"type":50,"value":654},"@clickhouse\u002Fclient-web",{"type":50,"value":188},{"type":44,"tag":190,"props":657,"children":659},{"id":658},"still-stuck",[660],{"type":50,"value":661},"Still Stuck?",{"type":44,"tag":82,"props":663,"children":664},{},[665,676,686],{"type":44,"tag":86,"props":666,"children":667},{},[668],{"type":44,"tag":109,"props":669,"children":673},{"href":670,"rel":671},"https:\u002F\u002Fclickhouse.com\u002Fdocs\u002Finterfaces\u002Fformats#rowbinary",[672],"nofollow",[674],{"type":50,"value":675},"ClickHouse RowBinary format",{"type":44,"tag":86,"props":677,"children":678},{},[679],{"type":44,"tag":109,"props":680,"children":683},{"href":681,"rel":682},"https:\u002F\u002Fclickhouse.com\u002Fdocs\u002Fsql-reference\u002Fdata-types",[672],[684],{"type":50,"value":685},"ClickHouse data types",{"type":44,"tag":86,"props":687,"children":688},{},[689],{"type":44,"tag":109,"props":690,"children":693},{"href":691,"rel":692},"https:\u002F\u002Fclickhouse.com\u002Fdocs\u002Fintegrations\u002Fjavascript",[672],[694],{"type":50,"value":695},"ClickHouse JS client docs",{"items":697,"total":888},[698,718,732,746,760,768,792,810,824,840,851,871],{"slug":699,"name":699,"fn":700,"description":701,"org":702,"tags":703,"stars":26,"repoUrl":27,"updatedAt":717},"chdb-datastore","accelerate pandas with chDB datastore","Use when the user has tabular data (pandas DataFrame, parquet, csv, Arrow, json) and wants to filter, group, aggregate, join, or speed up slow pandas. Provides chDB DataStore — same pandas API, ClickHouse engine underneath. Also handles reading from S3, MySQL, PostgreSQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake as DataFrames and joining across sources. TRIGGER when: user mentions DataFrame, parquet, csv, \"fast pandas\", \"speed up pandas\", or cross-source DataFrame joins; user imports `chdb.datastore` or `from datastore import DataStore`. SKIP this skill for raw SQL syntax (use chdb-sql instead), ClickHouse server administration, or non-Python DataStore API work.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[704,705,708,711,714],{"name":9,"slug":8,"type":13},{"name":706,"slug":707,"type":13},"Data Analysis","data-analysis",{"name":709,"slug":710,"type":13},"Data Engineering","data-engineering",{"name":712,"slug":713,"type":13},"Performance","performance",{"name":715,"slug":716,"type":13},"Python","python","2026-04-15T04:56:32.0629",{"slug":719,"name":719,"fn":720,"description":721,"org":722,"tags":723,"stars":26,"repoUrl":27,"updatedAt":731},"chdb-sql","run ClickHouse SQL queries in Python","Use when the user wants to run SQL — especially analytical SQL — on local files (parquet\u002Fcsv\u002Fjson), URLs, S3 paths, or remote databases (Postgres, MySQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake) without setting up a server. Provides chDB — embedded ClickHouse SQL in Python with 1000+ functions, Session for stateful multi-step pipelines, parametrized queries, and cross-source joins via `s3()`, `mysql()`, `postgresql()`, `iceberg()`, `deltaLake()`, `remoteSecure()` table functions. TRIGGER when: user wants SQL on parquet\u002Fcsv\u002Ffiles or across remote analytical sources; uses ClickHouse SQL features (window functions, windowFunnel, geoToH3, JSON path ops, Session, parametrized queries); imports `chdb` or calls `chdb.query()`. SKIP this skill for pandas-style DataFrame method-chaining (use chdb-datastore instead) or ClickHouse server administration.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[724,725,726,727,728],{"name":9,"slug":8,"type":13},{"name":706,"slug":707,"type":13},{"name":709,"slug":710,"type":13},{"name":715,"slug":716,"type":13},{"name":729,"slug":730,"type":13},"SQL","sql","2026-04-15T04:56:33.299509",{"slug":733,"name":733,"fn":734,"description":735,"org":736,"tags":737,"stars":26,"repoUrl":27,"updatedAt":745},"clickhouse-architecture-advisor","advise on ClickHouse architectures","MUST USE when designing ClickHouse architectures, selecting between ingestion or modeling patterns, or translating best practices into workload-specific system designs. Complements clickhouse-best-practices with decision frameworks and explicit provenance labels.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[738,741,742,743,744],{"name":739,"slug":740,"type":13},"Architecture","architecture",{"name":9,"slug":8,"type":13},{"name":709,"slug":710,"type":13},{"name":21,"slug":22,"type":13},{"name":712,"slug":713,"type":13},"2026-04-15T04:56:34.534001",{"slug":747,"name":747,"fn":748,"description":749,"org":750,"tags":751,"stars":26,"repoUrl":27,"updatedAt":759},"clickhouse-best-practices","review ClickHouse schemas and queries","MUST USE when reviewing ClickHouse schemas, queries, or configurations. Contains 31 rules that MUST be checked before providing recommendations. Always read relevant rule files and cite specific rules in responses.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[752,753,756,757,758],{"name":9,"slug":8,"type":13},{"name":754,"slug":755,"type":13},"Code Review","code-review",{"name":21,"slug":22,"type":13},{"name":712,"slug":713,"type":13},{"name":729,"slug":730,"type":13},"2026-04-06T18:07:18.953931",{"slug":4,"name":4,"fn":5,"description":6,"org":761,"tags":762,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[763,764,765,766,767],{"name":24,"slug":25,"type":13},{"name":9,"slug":8,"type":13},{"name":21,"slug":22,"type":13},{"name":18,"slug":19,"type":13},{"name":15,"slug":16,"type":13},{"slug":769,"name":769,"fn":770,"description":771,"org":772,"tags":773,"stars":26,"repoUrl":27,"updatedAt":791},"clickhouse-managed-postgres-rca","diagnose performance issues on ClickHouse-managed Postgres","MUST USE when investigating performance issues on a ClickHouse-managed Postgres instance. Provides an evidence-based RCA workflow that scrapes the Prometheus endpoint for system signal, pulls per-digest evidence from the Slow Query Patterns API, and recommends (does not apply) a fix.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[774,775,778,781,784,785,788],{"name":9,"slug":8,"type":13},{"name":776,"slug":777,"type":13},"Debugging","debugging",{"name":779,"slug":780,"type":13},"Monitoring","monitoring",{"name":782,"slug":783,"type":13},"Observability","observability",{"name":712,"slug":713,"type":13},{"name":786,"slug":787,"type":13},"PostgreSQL","postgresql",{"name":789,"slug":790,"type":13},"Prometheus","prometheus","2026-06-08T08:19:31.589978",{"slug":793,"name":793,"fn":794,"description":795,"org":796,"tags":797,"stars":26,"repoUrl":27,"updatedAt":809},"infra-clickhouse","manage ClickHouse server instances","Sets up and manages ClickHouse using the clickhousectl CLI — installs and runs a local ClickHouse server for development, and creates managed ClickHouse Cloud services for production (authentication, service creation, schema migration, application connection). Use when the user wants to build an application with ClickHouse, set up a local ClickHouse dev environment, create tables and start querying, deploy ClickHouse to production or ClickHouse Cloud, or migrate from a local setup to the cloud.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[798,801,802,803,806],{"name":799,"slug":800,"type":13},"CLI","cli",{"name":9,"slug":8,"type":13},{"name":21,"slug":22,"type":13},{"name":804,"slug":805,"type":13},"Deployment","deployment",{"name":807,"slug":808,"type":13},"Infrastructure","infrastructure","2026-07-28T06:06:00.02089",{"slug":811,"name":811,"fn":812,"description":813,"org":814,"tags":815,"stars":26,"repoUrl":27,"updatedAt":823},"infra-postgres","manage PostgreSQL database instances","Sets up and manages Postgres using the clickhousectl CLI — runs a local Docker-backed Postgres for development, and creates and operates managed ClickHouse Cloud Postgres services (connections, TLS, runtime config, read replicas, failover, point-in-time restore). Use when the user wants a Postgres or PostgreSQL database for their application, a local Postgres dev environment, psql access, or a managed\u002Fproduction Postgres in ClickHouse Cloud, or mentions moving a local Postgres to production.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[816,817,818,821,822],{"name":21,"slug":22,"type":13},{"name":804,"slug":805,"type":13},{"name":819,"slug":820,"type":13},"Docker","docker",{"name":807,"slug":808,"type":13},{"name":786,"slug":787,"type":13},"2026-07-28T06:06:07.151976",{"slug":425,"name":425,"fn":825,"description":826,"org":827,"tags":828,"stars":837,"repoUrl":838,"updatedAt":839},"build Node.js applications with ClickHouse","Write idiomatic application code with the ClickHouse Node.js client (`@clickhouse\u002Fclient`). Use this skill whenever a user is *building* against the Node.js client — configuring the client, pinging, inserting rows in JSON or raw formats, selecting and parsing results, binding query parameters, managing sessions and temporary tables, working with data types or customizing JSON parsing. Do NOT use for browser\u002FWeb client code.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[829,832,833,834],{"name":830,"slug":831,"type":13},"Backend","backend",{"name":9,"slug":8,"type":13},{"name":21,"slug":22,"type":13},{"name":835,"slug":836,"type":13},"Node.js","node-js",321,"https:\u002F\u002Fgithub.com\u002FClickHouse\u002Fclickhouse-js","2026-05-06T05:41:40.923776",{"slug":637,"name":637,"fn":841,"description":842,"org":843,"tags":844,"stars":837,"repoUrl":838,"updatedAt":850},"troubleshoot ClickHouse Node.js client","Troubleshoot and resolve common issues with the ClickHouse Node.js client (@clickhouse\u002Fclient). Use this skill whenever a user reports errors, unexpected behavior, or configuration questions involving the Node.js client specifically — including socket hang-up errors, Keep-Alive problems, stream handling issues, data type mismatches, read-only user restrictions, proxy\u002FTLS setup problems, or long-running query timeouts. Trigger even when the user hasn't precisely named the issue; vague symptoms like \"my inserts keep failing\" or \"connection drops randomly\" in a Node.js context are strong signals to use this skill. Do NOT use for browser\u002FWeb client issues.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[845,846,847,849],{"name":9,"slug":8,"type":13},{"name":776,"slug":777,"type":13},{"name":835,"slug":848,"type":13},"nodejs",{"name":15,"slug":16,"type":13},"2026-04-14T04:56:39.447406",{"slug":852,"name":852,"fn":853,"description":854,"org":855,"tags":856,"stars":868,"repoUrl":869,"updatedAt":870},"clickstack-otel-collector","set up OpenTelemetry collector for ClickHouse","Use when a user wants to wire an OpenTelemetry collector into a Managed ClickStack service on ClickHouse Cloud, either by deploying a new local collector (Docker run or Docker Compose) or by configuring their own existing collector, then send rich synthetic telemetry and verify it is visible in ClickStack.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[857,858,861,864,865],{"name":9,"slug":8,"type":13},{"name":859,"slug":860,"type":13},"Logs","logs",{"name":862,"slug":863,"type":13},"Metrics","metrics",{"name":782,"slug":783,"type":13},{"name":866,"slug":867,"type":13},"OpenTelemetry","opentelemetry",196,"https:\u002F\u002Fgithub.com\u002FClickHouse\u002Fclickhouse-docs","2026-06-11T08:25:17.959323",{"slug":872,"name":872,"fn":873,"description":874,"org":875,"tags":876,"stars":885,"repoUrl":886,"updatedAt":887},"setup","set up ClickHouse MCP server connections","Guides users through setting up the ClickHouse MCP server connection bundled with this plugin. Use when the user first installs the plugin or has trouble connecting to ClickHouse.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[877,878,879,882],{"name":9,"slug":8,"type":13},{"name":21,"slug":22,"type":13},{"name":880,"slug":881,"type":13},"Local Development","local-development",{"name":883,"slug":884,"type":13},"MCP","mcp",4,"https:\u002F\u002Fgithub.com\u002FClickHouse\u002Fclickhouse-claude-code-plugin","2026-04-19T04:59:29.849185",12,{"items":890,"total":949},[891,899,907,915,923,931,941],{"slug":699,"name":699,"fn":700,"description":701,"org":892,"tags":893,"stars":26,"repoUrl":27,"updatedAt":717},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[894,895,896,897,898],{"name":9,"slug":8,"type":13},{"name":706,"slug":707,"type":13},{"name":709,"slug":710,"type":13},{"name":712,"slug":713,"type":13},{"name":715,"slug":716,"type":13},{"slug":719,"name":719,"fn":720,"description":721,"org":900,"tags":901,"stars":26,"repoUrl":27,"updatedAt":731},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[902,903,904,905,906],{"name":9,"slug":8,"type":13},{"name":706,"slug":707,"type":13},{"name":709,"slug":710,"type":13},{"name":715,"slug":716,"type":13},{"name":729,"slug":730,"type":13},{"slug":733,"name":733,"fn":734,"description":735,"org":908,"tags":909,"stars":26,"repoUrl":27,"updatedAt":745},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[910,911,912,913,914],{"name":739,"slug":740,"type":13},{"name":9,"slug":8,"type":13},{"name":709,"slug":710,"type":13},{"name":21,"slug":22,"type":13},{"name":712,"slug":713,"type":13},{"slug":747,"name":747,"fn":748,"description":749,"org":916,"tags":917,"stars":26,"repoUrl":27,"updatedAt":759},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[918,919,920,921,922],{"name":9,"slug":8,"type":13},{"name":754,"slug":755,"type":13},{"name":21,"slug":22,"type":13},{"name":712,"slug":713,"type":13},{"name":729,"slug":730,"type":13},{"slug":4,"name":4,"fn":5,"description":6,"org":924,"tags":925,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[926,927,928,929,930],{"name":24,"slug":25,"type":13},{"name":9,"slug":8,"type":13},{"name":21,"slug":22,"type":13},{"name":18,"slug":19,"type":13},{"name":15,"slug":16,"type":13},{"slug":769,"name":769,"fn":770,"description":771,"org":932,"tags":933,"stars":26,"repoUrl":27,"updatedAt":791},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[934,935,936,937,938,939,940],{"name":9,"slug":8,"type":13},{"name":776,"slug":777,"type":13},{"name":779,"slug":780,"type":13},{"name":782,"slug":783,"type":13},{"name":712,"slug":713,"type":13},{"name":786,"slug":787,"type":13},{"name":789,"slug":790,"type":13},{"slug":793,"name":793,"fn":794,"description":795,"org":942,"tags":943,"stars":26,"repoUrl":27,"updatedAt":809},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[944,945,946,947,948],{"name":799,"slug":800,"type":13},{"name":9,"slug":8,"type":13},{"name":21,"slug":22,"type":13},{"name":804,"slug":805,"type":13},{"name":807,"slug":808,"type":13},8]