[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71RVSM279Y6VCQV0ZWWTDR0":3,"skill-same:01M71RVSM279Y6VCQV0ZWWTDR0":61},{"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":33,"linkOnly":34,"markdownContent":35,"markdownBody":36,"frontMatter":37,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":6,"ownerHtmlUrl":43,"ownerBlogUrl":44,"ownerDescription":45,"updatedAt":46,"install":47,"resources":60},"01M71RVSM279Y6VCQV0ZWWTDR0","matlas-search",null,"Use Matlas semantic search over published mathematical literature to find canonical statements, hypotheses, and citation metadata for standard or named results.","scaling-group\u002Feve","scaling-group","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F181708022?v=4","01M71RTM8RQW6CX3M1934895JA","Scientific Computing and Intelligence Group","01M71RTM99S28ZTFK9BK17R4GC",66,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fmatlas-search\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32],"search","matlas search","for","use","standard","literature","results","find","semantic search","canonical","Mathematical","semantic","metadata","Citation","2026-11-10 13:02:02",false,"---\nname: matlas-search\ndescription: \"Use Matlas semantic search over published mathematical literature to find canonical statements, hypotheses, and citation metadata for standard or named results.\"\n---\n\n# Matlas Search Manual\n\nUse this skill when a proof or evaluation needs a second retrieval source for:\n\n- a named theorem, lemma, proposition, corollary, definition, or standard fact;\n- a textbook-standard result that is hard to identify through TheoremSearch;\n- canonical citation metadata for a result from books or journal papers;\n- a check that a cited theorem's hypotheses match the current proof obligation.\n\nMatlas is supporting evidence only. It does not prove that a dependency is\nvalid for the current proof. Always compare the returned statement against the\nexact proof obligation.\n\nPublic references:\n\n- Web app: `https:\u002F\u002Fmatlas.ai\u002F`\n- API docs: `https:\u002F\u002Fmatlas.ai\u002Fdocs`\n- Search endpoint: `POST https:\u002F\u002Fmatlas.ai\u002Fapi\u002Fsearch`\n- Paper: `https:\u002F\u002Farxiv.org\u002Fabs\u002F2604.17484`\n\nMatlas searches published mathematical literature, including journal papers and\ntextbooks. It complements TheoremSearch: TheoremSearch is often better for\ntheorem-level statements from arXiv and structured web corpora; Matlas is often\nbetter for published or textbook sources with DOI-style provenance.\n\n## Basic Search\n\nUse natural language, LaTeX, or a mix. The API returns a JSON array of hits.\nRequest at least 10 results, then inspect the relevant top hits.\n\n```bash\ncurl -fsSL https:\u002F\u002Fmatlas.ai\u002Fapi\u002Fsearch \\\n  -H 'Content-Type: application\u002Fjson' \\\n  -d '{\n    \"query\": \"Banach fixed point theorem contraction complete metric space\",\n    \"num_results\": 10\n  }'\n```\n\nTypical hit fields include:\n\n```json\n{\n  \"type\": \"book\",\n  \"entity_name\": \"Theorem 2.12\",\n  \"doi\": \"10....\",\n  \"title\": \"An Introduction to Nonlinear Analysis\",\n  \"authors\": \"Martin Schechter\",\n  \"journal\": \"\",\n  \"year\": \"1991\",\n  \"statement\": \"Let ...\",\n  \"candidate_id\": \"...\"\n}\n```\n\nTrust the returned `statement` and source metadata more than the rank alone.\nA high-ranking hit can still have the wrong hypotheses, and a relevant hit can\nbe useful even if metadata such as DOI or year is missing.\n\n## Query Strategy\n\n1. State the needed result as a theorem: \"If [hypotheses], then [conclusion].\"\n2. Start with a broad query using standard terminology.\n3. Search again with distinctive words from a promising hit's `statement`,\n   `title`, or named theorem.\n4. Use Matlas alongside TheoremSearch when the dependency matters: agreement\n   across independent retrieval routes is stronger evidence than either alone.\n5. If the proof needs a citation, prefer hits with enough metadata to audit:\n   title, authors, journal or book, year, DOI, and exact statement.\n\nGood query patterns:\n\n- `\"spectral theorem compact self-adjoint operator Hilbert space\"`\n- `\"Nakayama lemma finitely generated module local ring\"`\n- `\"radical splitting theorem finite dimensional algebra\"`\n- `\"Frobenius reciprocity finite group representations\"`\n\nPoor query patterns:\n\n- `\"prove this\"`\n- proof paragraphs with irrelevant exposition;\n- local notation without mathematical context;\n- citation fishing when the proof has not stated the needed result.\n\n## Validation Checklist\n\nBefore using a Matlas hit in a proof or score, check:\n\n- objects match: groups, rings, schemes, spaces, modules, categories, fields;\n- hypotheses match: finiteness, completeness, compactness, smoothness,\n  noetherian assumptions, characteristic, topology, boundedness;\n- conclusion matches: existence, uniqueness, equivalence, vanishing, density,\n  openness, decomposition, exactness;\n- direction and variance match: source\u002Ftarget, left\u002Fright, covariant\u002F\n  contravariant, injection\u002Fsurjection;\n- notation imported from the source is compatible with the current proof;\n- no necessary condition is only present in surrounding text and absent from\n  the returned statement.\n\nIf a result is only close, treat it as a lead. Do not silently strengthen,\nweaken, or restate a theorem to fit the current proof.\n\n## Reporting Provenance\n\nWhen Matlas affects a proof step or evaluation, record enough for audit:\n\n```text\nMatlas: RESOLVED\nNeed: \u003Cproof obligation>\nStatement: \u003Cexact returned statement>\nSource: \u003Ctitle>, \u003Cauthors>, \u003Cjournal\u002Fbook>, \u003Cyear>, \u003Cdoi if present>\nCandidate id: \u003Ccandidate_id if present>\nUse: \u003Cwhy the hypotheses and conclusion match>\n```\n\nIf it is only partial:\n\n```text\nMatlas: PARTIAL\nNeed: \u003Cproof obligation>\nClosest statement: \u003Cexact returned statement>\nSource: \u003Cmetadata>\nMismatch: \u003Cspecific missing\u002Fextra hypothesis or wrong conclusion>\nNext query: \u003Csharper query>\n```\n\n## Failure Policy\n\nIf search fails:\n\n- Network, TLS, or API error: report `Matlas unavailable` with the endpoint and\n  error text.\n- Empty results: retry once with simpler language and fewer local symbols.\n- Low relevance: report `unresolved`, summarize the closest hit, and explain\n  the mismatch.\n- Missing DOI or metadata: use the available title\u002Fauthors\u002Fstatement, but say\n  what is missing.\n\nNever fabricate:\n\n- theorem names;\n- statement bodies;\n- authors;\n- titles;\n- DOI values;\n- years;\n- candidate IDs;\n- search scores.\n","# Matlas Search Manual\n\nUse this skill when a proof or evaluation needs a second retrieval source for:\n\n- a named theorem, lemma, proposition, corollary, definition, or standard fact;\n- a textbook-standard result that is hard to identify through TheoremSearch;\n- canonical citation metadata for a result from books or journal papers;\n- a check that a cited theorem's hypotheses match the current proof obligation.\n\nMatlas is supporting evidence only. It does not prove that a dependency is\nvalid for the current proof. Always compare the returned statement against the\nexact proof obligation.\n\nPublic references:\n\n- Web app: `https:\u002F\u002Fmatlas.ai\u002F`\n- API docs: `https:\u002F\u002Fmatlas.ai\u002Fdocs`\n- Search endpoint: `POST https:\u002F\u002Fmatlas.ai\u002Fapi\u002Fsearch`\n- Paper: `https:\u002F\u002Farxiv.org\u002Fabs\u002F2604.17484`\n\nMatlas searches published mathematical literature, including journal papers and\ntextbooks. It complements TheoremSearch: TheoremSearch is often better for\ntheorem-level statements from arXiv and structured web corpora; Matlas is often\nbetter for published or textbook sources with DOI-style provenance.\n\n## Basic Search\n\nUse natural language, LaTeX, or a mix. The API returns a JSON array of hits.\nRequest at least 10 results, then inspect the relevant top hits.\n\n```bash\ncurl -fsSL https:\u002F\u002Fmatlas.ai\u002Fapi\u002Fsearch \\\n  -H 'Content-Type: application\u002Fjson' \\\n  -d '{\n    \"query\": \"Banach fixed point theorem contraction complete metric space\",\n    \"num_results\": 10\n  }'\n```\n\nTypical hit fields include:\n\n```json\n{\n  \"type\": \"book\",\n  \"entity_name\": \"Theorem 2.12\",\n  \"doi\": \"10....\",\n  \"title\": \"An Introduction to Nonlinear Analysis\",\n  \"authors\": \"Martin Schechter\",\n  \"journal\": \"\",\n  \"year\": \"1991\",\n  \"statement\": \"Let ...\",\n  \"candidate_id\": \"...\"\n}\n```\n\nTrust the returned `statement` and source metadata more than the rank alone.\nA high-ranking hit can still have the wrong hypotheses, and a relevant hit can\nbe useful even if metadata such as DOI or year is missing.\n\n## Query Strategy\n\n1. State the needed result as a theorem: \"If [hypotheses], then [conclusion].\"\n2. Start with a broad query using standard terminology.\n3. Search again with distinctive words from a promising hit's `statement`,\n   `title`, or named theorem.\n4. Use Matlas alongside TheoremSearch when the dependency matters: agreement\n   across independent retrieval routes is stronger evidence than either alone.\n5. If the proof needs a citation, prefer hits with enough metadata to audit:\n   title, authors, journal or book, year, DOI, and exact statement.\n\nGood query patterns:\n\n- `\"spectral theorem compact self-adjoint operator Hilbert space\"`\n- `\"Nakayama lemma finitely generated module local ring\"`\n- `\"radical splitting theorem finite dimensional algebra\"`\n- `\"Frobenius reciprocity finite group representations\"`\n\nPoor query patterns:\n\n- `\"prove this\"`\n- proof paragraphs with irrelevant exposition;\n- local notation without mathematical context;\n- citation fishing when the proof has not stated the needed result.\n\n## Validation Checklist\n\nBefore using a Matlas hit in a proof or score, check:\n\n- objects match: groups, rings, schemes, spaces, modules, categories, fields;\n- hypotheses match: finiteness, completeness, compactness, smoothness,\n  noetherian assumptions, characteristic, topology, boundedness;\n- conclusion matches: existence, uniqueness, equivalence, vanishing, density,\n  openness, decomposition, exactness;\n- direction and variance match: source\u002Ftarget, left\u002Fright, covariant\u002F\n  contravariant, injection\u002Fsurjection;\n- notation imported from the source is compatible with the current proof;\n- no necessary condition is only present in surrounding text and absent from\n  the returned statement.\n\nIf a result is only close, treat it as a lead. Do not silently strengthen,\nweaken, or restate a theorem to fit the current proof.\n\n## Reporting Provenance\n\nWhen Matlas affects a proof step or evaluation, record enough for audit:\n\n```text\nMatlas: RESOLVED\nNeed: \u003Cproof obligation>\nStatement: \u003Cexact returned statement>\nSource: \u003Ctitle>, \u003Cauthors>, \u003Cjournal\u002Fbook>, \u003Cyear>, \u003Cdoi if present>\nCandidate id: \u003Ccandidate_id if present>\nUse: \u003Cwhy the hypotheses and conclusion match>\n```\n\nIf it is only partial:\n\n```text\nMatlas: PARTIAL\nNeed: \u003Cproof obligation>\nClosest statement: \u003Cexact returned statement>\nSource: \u003Cmetadata>\nMismatch: \u003Cspecific missing\u002Fextra hypothesis or wrong conclusion>\nNext query: \u003Csharper query>\n```\n\n## Failure Policy\n\nIf search fails:\n\n- Network, TLS, or API error: report `Matlas unavailable` with the endpoint and\n  error text.\n- Empty results: retry once with simpler language and fewer local symbols.\n- Low relevance: report `unresolved`, summarize the closest hit, and explain\n  the mismatch.\n- Missing DOI or metadata: use the available title\u002Fauthors\u002Fstatement, but say\n  what is missing.\n\nNever fabricate:\n\n- theorem names;\n- statement bodies;\n- authors;\n- titles;\n- DOI values;\n- years;\n- candidate IDs;\n- search scores.\n",{"name":5,"description":7},"main","979065fd2cf1b3be9caf70596478fe27df7815ad","docs\u002Fskills\u002Fmatlas-search",9,"https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve","https:\u002F\u002Fgithub.com\u002Fscaling-group","https:\u002F\u002Fscaling-group.github.io\u002F","Department of Mathematics, National University of Singapore","2026-08-20 15:35:50",{"source":8,"skillName":5,"commands":48,"prompt":58,"zipAvailable":59},[49,52,55],{"tool":50,"command":51},"npx","npx skills add scaling-group\u002Feve --skill matlas-search",{"tool":53,"command":54},"bunx","bunx skills add scaling-group\u002Feve --skill matlas-search",{"tool":56,"command":57},"pnpm","pnpm dlx skills add scaling-group\u002Feve --skill matlas-search","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fmatlas-search\u002FSKILL.md",true,[],[62,87,101,116,135,150],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":78,"publishedAt":33},"01M71RW4VG8A6FFRGT7H41J16J","writing-skills","Use when creating new skills, editing existing skills, or verifying skills work before deployment","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fwriting-skills\u002FSKILL.md",[68,73],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":74,"categoryName":75,"taxonomyType":76,"taxonomyTypeDesc":77},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[79,80,81,82,83,84,22,85,86],"writing","writing skills","work","editing","deployment","new","Existing","when",{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":96,"publishedAt":33},"01M71RW2PFRVWH24AP25T6AV4E","theorem-search","Use TheoremSearch to find, verify, and cite mathematical theorem statements, named results, proof routes, and literature dependencies.","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Ftheorem-search\u002FSKILL.md",[93],{"categoryNo":94,"categoryName":95,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000021","资料检索与研究",[97,19,22,24,25,98,26,99,100,29],"theorem search","verify","dependencies","proof",{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":33},"01M71RW0N0GQR9GW1B3HKBB800","supervise-run","Use when monitoring a running Eve experiment and recovering from problems while it runs","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fsupervise-run\u002FSKILL.md",[],[108,109,110,111,112,113,114,86,22,115],"supervise run","supervise","run","monitoring","experiment","it","running","eve",{"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":122,"publishedAt":33},"01M71RVYA101VVX3CKR2HZHBT3","run-math-proof-quickstart","Use when running a Math Proof quickstart from a user-provided problem: create a local task copy, launch a real EvE attempt, supervise it, and report proof\u002Fevaluation results.","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Frun-math-proof-quickstart\u002FSKILL.md",[],[110,123,124,125,126,100,127,128,114,129,130,115,22,86,131,132,25,109,133,134],"math","run math proof quickstart","quickstart","math proof","copy","launch","problem","local","task","create","report","Evaluation",{"skillNo":136,"name":137,"nameZh":6,"description":138,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":139,"categories":140,"keywords":141,"publishedAt":33},"01M71RVWYKH6TF0WB6X57GMWJ2","run-circle-packing-smoke","Use when a fast end-to-end smoke run is needed for the Eve loop development, will run on circle packing task.","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Frun-circle-packing-smoke\u002FSKILL.md",[],[142,143,110,144,145,86,146,21,147,115,22,148,131,149],"circle","packing","smoke","run circle packing smoke","development","end","loop","fast",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":159,"publishedAt":33},"01M71RVVCE1NEF4ZM405WX815Z","resume-run","Use when continuing an interrupted or paused Eve run from where it stopped (same run, same id), optionally after editing the experiment","https:\u002F\u002Fgithub.com\u002Fscaling-group\u002Feve\u002Fblob\u002Fmain\u002Fdocs\u002Fskills\u002Fresume-run\u002FSKILL.md",[156],{"categoryNo":157,"categoryName":158,"taxonomyType":76,"taxonomyTypeDesc":77},"01JBSEEDCAT000000000000038","人力资源",[160,110,161,82,115,22,86,162,163,112,113],"resume","resume run","id","same"]