[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DBM0873JSWD7M10C20MK2R":3,"skill-same:01M6DBM0873JSWD7M10C20MK2R":64},{"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":14,"repositoryHtmlUrl":47,"repositoryHomepageUrl":6,"ownerHtmlUrl":48,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":49,"install":50,"resources":63},"01M6DBM0873JSWD7M10C20MK2R","multilingual-search",null,"Reach non-English sources deliberately - query in the target language against a native-index engine (Yandex for RU, Quark\u002F360search for CN, Naver for KR, Seznam for CZ), fuse several engines with Reciprocal Rank Fusion, and cite the original while translating only on output. Use when a research question has regional, non-English, or non-Western coverage that an English-only search will miss, when results feel thin or one-sided, when a topic is primarily discussed in Russian\u002FChinese\u002FKorean\u002FJapanese\u002FArabic sources, or when a claim needs corroboration from an index Google and Bing do not dominate. Also use when merging results from several search providers into one ranking.","Prekzursil\u002Fagent-skills-toolchain","Prekzursil","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54636077?v=4","01M6DACZ1PCQHD99SCR1YXG28B","Andrei Mihai Visalon","01M6DACZ670WZF535V3WYYSS4Z",0,"MIT","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fmultilingual-search\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"multilingual search","search","multilingual","japanese","fusion","claim","query","do","russian","arabic","korean","chinese","google","only","use","coverage","results","question","engine","research","2026-11-10 10:12:30",false,"---\nname: multilingual-search\ndescription: Reach non-English sources deliberately - query in the target language against a native-index engine (Yandex for RU, Quark\u002F360search for CN, Naver for KR, Seznam for CZ), fuse several engines with Reciprocal Rank Fusion, and cite the original while translating only on output. Use when a research question has regional, non-English, or non-Western coverage that an English-only search will miss, when results feel thin or one-sided, when a topic is primarily discussed in Russian\u002FChinese\u002FKorean\u002FJapanese\u002FArabic sources, or when a claim needs corroboration from an index Google and Bing do not dominate. Also use when merging results from several search providers into one ranking.\n---\n\n# Multilingual search\n\nEnglish-first search does not return a translated view of the world's sources — it returns\nthe subset that English-language indexes cover. For anything regional (domestic industry,\nlocal regulation, national politics, area-studies scholarship, a product sold in one\nmarket), the highest-value documents may have no English equivalent at all. Reaching them\ntakes three deliberate moves.\n\n## The three rules\n\n**1. Query in the target language.** A translated *result* is useful; a translated *query*\nis a different query. `国产大飞机` and \"Chinese domestic airliner\" retrieve different\ndocuments from different indexes — the Chinese phrase is what Chinese authors actually\nwrote. Do not machine-translate word-for-word: use the term a native writer would use.\nWhen unsure, search the English term first, find the native-language phrase in a bilingual\npage (Wikipedia interlanguage links are reliable for this), then re-query with it.\n\n**2. Cite the original, translate only on output.** The citation is the native-language URL\nand its native-language title. Quote the original text, then give your translation beside\nit. Never present a translation as if it were the source's own words, and never cite an\nEnglish article *about* a native source when you have the native source.\n\n**3. Route the query at an index that actually covers the region.** Google and DuckDuckGo\nunder-index Yandex-, Baidu-, and Naver-native content. Use a native engine.\n\n## Routing table — measured 2026-07-24 against the local SearXNG (`127.0.0.1:8890`)\n\n| Need | Engine | Bang | Notes |\n|---|---|---|---|\n| Russian | `yandex` | `!yd` | best RU coverage; strong on Habr, Dzen, RU academia |\n| Chinese | `quark` | `!qk` | Alibaba's engine; most reliable CN route measured |\n| Chinese (2nd opinion) | `360search` | `!360so` | thinner but independent of Quark |\n| Chinese dev\u002FIT | `baidu kaifa` | `!bdk` | Baidu's developer vertical — live even though general Baidu is not |\n| Korean news | `naver news` | `!nvrn` | `categories=news` only; the general `naver` engine is broken |\n| Czech | `seznam` | `!szn` | strong native index |\n| Independent crawl | `mojeek` | `!mjk` | own crawler — genuinely independent of the big four |\n\n**Blocked, do not rely on** (measured): `baidu` (CAPTCHA), `sogou` (CAPTCHA\u002Ftimeout),\n`naver` general (returns 0 with no error), `yahoo`, `qwant`.\n\nThis table rots — engines add CAPTCHAs and change markup. Re-probe before trusting it:\n\n```bash\npython tools\u002Fsearch-expansion\u002Fsearxng_regional.py --probe\n```\n\n## How to actually issue the query\n\nNo configuration is needed. A SearXNG engine marked `disabled` is still reachable on\ndemand — `disabled` only governs the *default* result mix:\n\n```\nGET http:\u002F\u002F127.0.0.1:8890\u002Fsearch?q=\u003Cquery>&engines=yandex&format=json\nGET http:\u002F\u002F127.0.0.1:8890\u002Fsearch?q=!yd \u003Cquery>&format=json\nGET http:\u002F\u002F127.0.0.1:8890\u002Fsearch?q=\u003Cquery>&engines=quark,360search&language=zh-CN&format=json\n```\n\n`language=` is honoured and changes ranking, not just filtering (measured: `ru-RU`\npromoted a Habr article over the Wikipedia hit that led the unlocalised run). Several\nengines can be addressed in one request, and results carry per-engine attribution.\n\nVia the MCP surface, `mcp__searxng__searxng_web_search` reaches the same instance.\n\n## Two failure shapes that read as success\n\n* **Silent zero.** An engine can return `n=0` with an **empty** `unresponsive_engines`\n  list — no error to notice. `naver` does exactly this while `naver news` works. Assert on\n  result *count*, never on absence of an error.\n* **Sticky suspension.** After a CAPTCHA, SearXNG suspends the engine and later calls\n  return in ~0.01 s with `Suspended: CAPTCHA`. An instant empty reply is a suspension, not\n  an answer; retrying immediately proves nothing.\n\n## Fuse, do not concatenate\n\nSeveral engines return overlapping, differently-ordered lists. Merge them with Reciprocal\nRank Fusion rather than stacking or trusting one engine's order:\n\n```python\nfrom rank_fusion import fuse_provider_payloads  # tools\u002Fsearch-expansion\u002F\n\nfused = fuse_provider_payloads(\n    {\"searxng-ru\": ru_payload, \"searxng-cn\": cn_payload, \"exa\": exa_payload},\n    limit=20,\n)\nfor r in fused:\n    print(r.score, r.sources, r.display_url)\n```\n\nRRF uses only each provider's *rank ordinal*, never its relevance score — those scores are\nnot comparable across providers, and normalising them invents a shared scale. A document\nseveral independent engines rank highly outranks one engine's lone top hit, which is\nexactly the cross-index corroboration multilingual search is for. `sources` on each result\ntells you which indexes agreed.\n\n## Reporting\n\nState which languages and engines you searched, in the report. \"I found nothing\" is only\nmeaningful if you say where you looked — an English-only sweep that missed a Russian\nliterature is a *coverage* result, not an *absence* result. If a claim rests on a single\nregional source with no independent corroboration, say so and give the confidence band.\n","# Multilingual search\n\nEnglish-first search does not return a translated view of the world's sources — it returns\nthe subset that English-language indexes cover. For anything regional (domestic industry,\nlocal regulation, national politics, area-studies scholarship, a product sold in one\nmarket), the highest-value documents may have no English equivalent at all. Reaching them\ntakes three deliberate moves.\n\n## The three rules\n\n**1. Query in the target language.** A translated *result* is useful; a translated *query*\nis a different query. `国产大飞机` and \"Chinese domestic airliner\" retrieve different\ndocuments from different indexes — the Chinese phrase is what Chinese authors actually\nwrote. Do not machine-translate word-for-word: use the term a native writer would use.\nWhen unsure, search the English term first, find the native-language phrase in a bilingual\npage (Wikipedia interlanguage links are reliable for this), then re-query with it.\n\n**2. Cite the original, translate only on output.** The citation is the native-language URL\nand its native-language title. Quote the original text, then give your translation beside\nit. Never present a translation as if it were the source's own words, and never cite an\nEnglish article *about* a native source when you have the native source.\n\n**3. Route the query at an index that actually covers the region.** Google and DuckDuckGo\nunder-index Yandex-, Baidu-, and Naver-native content. Use a native engine.\n\n## Routing table — measured 2026-07-24 against the local SearXNG (`127.0.0.1:8890`)\n\n| Need | Engine | Bang | Notes |\n|---|---|---|---|\n| Russian | `yandex` | `!yd` | best RU coverage; strong on Habr, Dzen, RU academia |\n| Chinese | `quark` | `!qk` | Alibaba's engine; most reliable CN route measured |\n| Chinese (2nd opinion) | `360search` | `!360so` | thinner but independent of Quark |\n| Chinese dev\u002FIT | `baidu kaifa` | `!bdk` | Baidu's developer vertical — live even though general Baidu is not |\n| Korean news | `naver news` | `!nvrn` | `categories=news` only; the general `naver` engine is broken |\n| Czech | `seznam` | `!szn` | strong native index |\n| Independent crawl | `mojeek` | `!mjk` | own crawler — genuinely independent of the big four |\n\n**Blocked, do not rely on** (measured): `baidu` (CAPTCHA), `sogou` (CAPTCHA\u002Ftimeout),\n`naver` general (returns 0 with no error), `yahoo`, `qwant`.\n\nThis table rots — engines add CAPTCHAs and change markup. Re-probe before trusting it:\n\n```bash\npython tools\u002Fsearch-expansion\u002Fsearxng_regional.py --probe\n```\n\n## How to actually issue the query\n\nNo configuration is needed. A SearXNG engine marked `disabled` is still reachable on\ndemand — `disabled` only governs the *default* result mix:\n\n```\nGET http:\u002F\u002F127.0.0.1:8890\u002Fsearch?q=\u003Cquery>&engines=yandex&format=json\nGET http:\u002F\u002F127.0.0.1:8890\u002Fsearch?q=!yd \u003Cquery>&format=json\nGET http:\u002F\u002F127.0.0.1:8890\u002Fsearch?q=\u003Cquery>&engines=quark,360search&language=zh-CN&format=json\n```\n\n`language=` is honoured and changes ranking, not just filtering (measured: `ru-RU`\npromoted a Habr article over the Wikipedia hit that led the unlocalised run). Several\nengines can be addressed in one request, and results carry per-engine attribution.\n\nVia the MCP surface, `mcp__searxng__searxng_web_search` reaches the same instance.\n\n## Two failure shapes that read as success\n\n* **Silent zero.** An engine can return `n=0` with an **empty** `unresponsive_engines`\n  list — no error to notice. `naver` does exactly this while `naver news` works. Assert on\n  result *count*, never on absence of an error.\n* **Sticky suspension.** After a CAPTCHA, SearXNG suspends the engine and later calls\n  return in ~0.01 s with `Suspended: CAPTCHA`. An instant empty reply is a suspension, not\n  an answer; retrying immediately proves nothing.\n\n## Fuse, do not concatenate\n\nSeveral engines return overlapping, differently-ordered lists. Merge them with Reciprocal\nRank Fusion rather than stacking or trusting one engine's order:\n\n```python\nfrom rank_fusion import fuse_provider_payloads  # tools\u002Fsearch-expansion\u002F\n\nfused = fuse_provider_payloads(\n    {\"searxng-ru\": ru_payload, \"searxng-cn\": cn_payload, \"exa\": exa_payload},\n    limit=20,\n)\nfor r in fused:\n    print(r.score, r.sources, r.display_url)\n```\n\nRRF uses only each provider's *rank ordinal*, never its relevance score — those scores are\nnot comparable across providers, and normalising them invents a shared scale. A document\nseveral independent engines rank highly outranks one engine's lone top hit, which is\nexactly the cross-index corroboration multilingual search is for. `sources` on each result\ntells you which indexes agreed.\n\n## Reporting\n\nState which languages and engines you searched, in the report. \"I found nothing\" is only\nmeaningful if you say where you looked — an English-only sweep that missed a Russian\nliterature is a *coverage* result, not an *absence* result. If a claim rests on a single\nregional source with no independent corroboration, say so and give the confidence band.\n",{"name":5,"description":7},"main","a298c929b6cbd662e1fce54302ce520d5ceea1a5",".agents\u002Fskills\u002Fmultilingual-search","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain","https:\u002F\u002Fgithub.com\u002FPrekzursil","2026-11-02 14:30:57",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add Prekzursil\u002Fagent-skills-toolchain --skill multilingual-search",{"tool":56,"command":57},"bunx","bunx skills add Prekzursil\u002Fagent-skills-toolchain --skill multilingual-search",{"tool":59,"command":60},"pnpm","pnpm dlx skills add Prekzursil\u002Fagent-skills-toolchain --skill multilingual-search","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fmultilingual-search\u002FSKILL.md",true,[],[65,86,116,143,157,173],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":39},"01M6DF7D7478R55P46REYH4QAC","clarify-requirements","Resolve genuine pre-execution ambiguity by scanning repo context, generating 1-6 grounded options, and asking 1-3 focused questions via AskUserQuestion. NOT for mid-task use (injection risk); NOT when the goal is already unambiguous; use to disambiguate vague prompts before any file edits.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fworkspace-hooks\u002Fskills\u002Fclarify-requirements\u002FSKILL.md",[],[72,73,74,75,76,77,78,79,80,33,81,82,83,84,85],"clarify requirements","clarify","requirements","risk","options","task","prompts","goal","scanning","context","repo","injection","execution","file",{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":97,"publishedAt":39},"01M6DF79CA4BHJS66J8EZDBPWJ","openmm-md","Run FREE, open-source, Windows-native GPU molecular dynamics via the openmm-md MCP (OpenMM + CUDA on the local NVIDIA GPU). Use for protein \u002F protein-ligand MD when you want a free stack or are on Windows without a Schrodinger license. Covers the full workflow (prepare, solvate, minimize, equilibrate, production, analyze, mmgbsa) AND an honest decision guide vs Schrodinger Desmond, including how to run Desmond legitimately on Linux or a cloud GPU with a real license. This is a deliberately LESSER free alternative and does NOT replace the schrodinger MCP tools. Do NOT use it to pick a free tool when a validated FEP+, WaterMap, IFD-MD, or Prime MM-GBSA result is actually required.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fmcp\u002Fopenmm-md-mcp\u002Fskill\u002Fopenmm-md\u002FSKILL.md",[92],{"categoryNo":93,"categoryName":94,"taxonomyType":95,"taxonomyTypeDesc":96},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[98,99,100,101,102,103,104,105,106,107,108,109,110,111,33,112,113,114,26,115],"openmm md","how","decision","pick","linux","analyze","tool","stack","open","NVIDIA","gpu","license","workflow","windows","free","molecular dynamics","local","MCP",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":128,"publishedAt":39},"01M6DF76C1MG85VEWP8VZCBNMN","visual-review","Take screenshots of web pages and UI using Playwright for visual review and iteration","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fvisual-review\u002FSKILL.md",[122,125],{"categoryNo":123,"categoryName":124,"taxonomyType":95,"taxonomyTypeDesc":96},"01JBSEEDCAT000000000000003","演示文稿",{"categoryNo":126,"categoryName":127,"taxonomyType":95,"taxonomyTypeDesc":96},"01JBSEEDCAT000000000000008","测试与质量",[129,130,131,132,133,134,135,136,137,138,99,139,140,141,142],"review","visual","visual review","pages","web","false","iteration","take","check","slides","ui","screenshot","playwright","screenshots",{"skillNo":144,"name":145,"nameZh":6,"description":146,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":147,"categories":148,"keywords":149,"publishedAt":39},"01M6DF73AQVCBZKEN0YXEX9FXP","status","Diagnostic status report — shows metaswarm installation state, project setup, and potential issues","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstatus\u002FSKILL.md",[],[145,150,151,152,153,154,155,156],"setup","report","project","installation","project setup","issues","state",{"skillNo":158,"name":159,"nameZh":6,"description":160,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":161,"categories":162,"keywords":164,"publishedAt":39},"01M6DF6X6C2XB0P6GX76P7TTYV","start","Use when starting work on any task, when the user mentions metaswarm, or when the user wants to begin tracked development work","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstart\u002FSKILL.md",[163],{"categoryNo":93,"categoryName":94,"taxonomyType":95,"taxonomyTypeDesc":96},[159,165,77,166,167,168,33,169,170,171,172],"start issue","issue","ready","work on issue","user","work","work on","development",{"skillNo":174,"name":150,"nameZh":6,"description":175,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":176,"categories":177,"keywords":178,"publishedAt":39},"01M6DF6QMJ9AJKFKHHKRT005PQ","Interactive project setup — detects your project, configures metaswarm, writes project-local files","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fsetup\u002FSKILL.md",[],[150,152,114,179,154,180],"files","interactive"]