[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"creator:01M7DEC83S24HX3V8G1JSWKHD0":3,"creator-list:01M7DEC83S24HX3V8G1JSWKHD0:stars:1:":20},{"accountNo":4,"login":5,"displayName":6,"avatarUrl":7,"accountType":8,"accountTypeDesc":9,"profileDescription":10,"skillCount":11,"repoCount":12,"professionCount":13,"scenarioCount":13,"totalStars":13,"updatedAt":14,"htmlUrl":15,"blogUrl":16,"company":6,"location":17,"followersCount":18,"publicReposCount":19},"01M7DEC83S24HX3V8G1JSWKHD0","szl-holdings","SZL Holdings","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F273360816?v=4","ORGANIZATION","组织","Governed AI infrastructure for inspectable decisions: source, runtime state, receipts, and proof boundaries.",47,1,0,"2026-11-15 03:52:34","https:\u002F\u002Fgithub.com\u002Fszl-holdings","https:\u002F\u002Fa-11-oy.com","New York, NY",5,134,{"records":21,"total":11,"current":12,"size":150},[22,34,41,48,54,59,66,73,80,87,94,101,108,115,122,129,136,143],{"skillNo":23,"name":24,"nameZh":25,"description":26,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":30,"categories":31,"keywords":32,"publishedAt":33},"01M7DENG22R761RTDSG11WWJB3","szl-unit-invariants",null,"Checks declared SI dimensions, allowlisted unit conversions, expected ranges and fixed pointwise invariants (equal, sum, difference, product, ratio) over selected measurement or simulation columns, flagging incompatible quantities, scale errors of 1000x and broken conservation checks. Use when merging instrument exports with different units, validating simulation output against conservation laws, or when a reviewer asks how unit errors were ruled out. Not an uncertainty-propagation or symbolic tool.","szl-holdings\u002Fszl-skills","01M7DEC85084EGHEJH2KMAA53K","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-unit-invariants\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":35,"name":36,"nameZh":25,"description":37,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":38,"categories":39,"keywords":40,"publishedAt":33},"01M7DENAVBJ6GSCRT3E1PWK56D","szl-uncertainty-lineage","Propagate declared measurement uncertainty through a small arithmetic dependency graph using a first-order delta approximation. Use when a researcher needs to see how corrected inputs or shared-source correlation change a reported standard uncertainty; require unit checks separately.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-uncertainty-lineage\u002FSKILL.md",[],[],{"skillNo":42,"name":43,"nameZh":25,"description":44,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":45,"categories":46,"keywords":47,"publishedAt":33},"01M7DEN2M7P59RTM6RJE99PGN8","szl-typesafe-ai","TypeSafe\u002FJev: Choice\u002FNoul\u002FScore in code, not TypeScript types. System One models turn natural language and application state into probabilities that software can compose. Use when building routing, ranking, extraction, verification, or replacing an LLM prompt-and-parse step with structured decisions; also when brainstorming what TypeSafe could make possible. Do not use for TypeScript\u002FJava\u002FRust static types, Zod\u002FPydantic\u002Fmypy\u002FJSON Schema\u002Fprotobuf codegen, or asking whether a pipeline is type-safe. At SZL, Jev is an optional second reader only.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-typesafe-ai\u002FSKILL.md",[],[],{"skillNo":49,"name":50,"nameZh":25,"description":25,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":33},"01M7DEN1C9G4H12N9HRHYK2HZ7","demo","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-skill-update-review\u002Fassets\u002Fexample\u002Fold-package\u002Fskills\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":55,"name":50,"nameZh":25,"description":25,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":33},"01M7DEMZYRXT5CG2XPXERAM218","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-skill-update-review\u002Fassets\u002Fexample\u002Fnew-package\u002Fskills\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":25,"description":62,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":33},"01M7DEMVH2NMG4PS0DZTMGATAF","szl-skill-update-review","Compare pinned local skill packages using inventories and a retained lock; report changed skills, files, declarations and reruns without approving safety or validity.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-skill-update-review\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":25,"description":69,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":33},"01M7DEMKV29EMH203F3Y1WGD99","szl-skill-conformance","Checks retained scientific skill-sharing evidence: exact local source and supplied host-export resource bytes, explicit import aliases, task and result digests, positive versus irrelevant-task trigger declarations, and optional artifact-to-trace byte bindings. Use when a shared skill appears imported but invocation is uncertain, when comparing retained trials across hosts, or when an update may invalidate an earlier result. Offline checking never executes the candidate and never certifies real host invocation or scientific correctness.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-skill-conformance\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":25,"description":76,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":33},"01M7DEMC43TC2W8YEBYY2JC1SH","szl-session-receipt","Writes a verifiable record of an analysis session: SHA-256 of every input, script and output file, the commands and environment as declared, a root digest, and a plain-language Methods paragraph; later re-hashes the files to report MATCH \u002F MISMATCH \u002F MISSING. Use at the end of a Claude Science or notebook session, when handing an analysis to a collaborator, when a journal asks for a reproducibility statement, or when the user says 'record what we just did'. Not a signature of correctness.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-session-receipt\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":25,"description":83,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":33},"01M7DEM4N77MPHZTVEYKF5ZA45","szl-science-workbench","Runs a resumable local project with selected dataset, binary\u002Fcategorical model, numerical math, kernel, calibration-benchmark and paired checks; retains an anatomy graph, byte hashes and immutable capsules. Use to run these supported checks together, resume a project or determine which dependent conclusions changed. Other science-pack tools run separately through their own CLIs. Not an experiment runner, model trainer or dispatcher for every science skill.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-science-workbench\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":25,"description":90,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":33},"01M7DEKXVP9513QET9PY0F36TT","szl-reviewer-pack","Renders one REVIEW.md a reviewer can read in five minutes from a szl-science-workbench project (latest run, every check's status, report integrity, each unresolved finding quoted) plus any standalone reports (evidence gate, cross-implementation check, mutation coverage, energy receipt, session receipt), with an optional reviewer-declared non-compensatory aggregate. Use when the user asks for a pre-review, a PI summary of what was checked, a response-to-reviewers appendix, or 'what is still open before we submit'. Not an approval to publish.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-reviewer-pack\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":25,"description":97,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":33},"01M7DEKP3JK3J1EPF9WTB02CSZ","szl-retrieval-eval","Scores a literature or document retrieval system against relevance judgments: nDCG@k, MRR, MAP, precision@k and recall@k per query and on average, with judged queries that the system never answered kept in the denominator at zero, unjudged queries listed rather than silently scored, and duplicate results dropped and counted. Accepts JSON or TREC run\u002Fqrels files. Use when a lab evaluates a RAG or search pipeline over papers, protocols or records, compares two retrievers, or asks 'did the new embedding model actually find more of the relevant papers'. Not a judgment of whether the relevance labels are right.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-retrieval-eval\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":25,"description":104,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":33},"01M7DEKFBGM5V6GTVVP2HFC1NR","szl-result-fragility","Computes the fragility index of a two-arm binary result: Fisher's exact p on the 2x2 table, the smallest number of participants whose outcome would have to change for significance to disappear, the fragility quotient, and a FRAGILE \u002F ROBUST judgment against the number lost to follow-up; for non-significant results, the reverse fragility index. Exact arithmetic, stdlib only. Use when a trial or A\u002FB result 'reached significance' on a handful of events, when a reviewer asks how fragile a finding is, when comparing a result with its dropout count, or when the user says 'how many patients away from p = 0.05 is this'. Not an effect-size or design assessment.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-result-fragility\u002FSKILL.md",[],[],{"skillNo":109,"name":110,"nameZh":25,"description":111,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":33},"01M7DEK8MN490BY7NZ4MPMH1QX","szl-research-change-impact","Compares two retained research dependency snapshots and supplied artifact readbacks, preserving removed nodes and edges when tracing impact. Use before replacing a dataset, model, kernel or analysis plan, or when a graph edit could hide which claims relied on withdrawn evidence. Produces a deterministic recheck order and witness paths; does not execute experiments or judge scientific truth.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-research-change-impact\u002FSKILL.md",[],[],{"skillNo":116,"name":117,"nameZh":25,"description":118,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":119,"categories":120,"keywords":121,"publishedAt":33},"01M7DEK1FN7PY62B5ZNV9R98ER","szl-research-anatomy","Maintains a project's claim-and-evidence ledger as a dependency graph: datasets, code, runs, claims and proofs with supporting, contradicting and qualifying evidence links, observation dates, expiry, source revisions and corrections, then reports which conclusions need rechecking when any input changes. Use for persistent project memory across sessions, when a source is retracted or updated, when two findings conflict, or when the user asks 'what depends on this'. Not a literature monitor and not a truth oracle.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-research-anatomy\u002FSKILL.md",[],[],{"skillNo":123,"name":124,"nameZh":25,"description":125,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":126,"categories":127,"keywords":128,"publishedAt":33},"01M7DEJSYXNWYJK4W8V4T7YNT1","szl-reproducibility-capsule","Creates an exact-byte manifest (path, size, SHA-256) for the explicitly selected files of an experiment, verifies them later as changed \u002F missing \u002F unsafe, and validates a bounded replay declaration (inputs, source, environment, plan, outputs) without executing anything. Use when handing an analysis to a collaborator, archiving a result, preparing a replication or when the user asks 'did anything change since the run'. Not a container, not a signature, and it does not reproduce the experiment.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-reproducibility-capsule\u002FSKILL.md",[],[],{"skillNo":130,"name":131,"nameZh":25,"description":132,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":33},"01M7DEJNCHJM11S0XT4STVBZ3E","szl-reporting-guideline-audit","Map a manuscript onto the frozen PRISMA 2020, CONSORT 2025, or STROBE 2007 checklist and show every omitted item. Use before calling a paper complete against one of those guidelines. A locator is not visual confirmation, and the report does not judge reporting compliance.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-reporting-guideline-audit\u002FSKILL.md",[],[],{"skillNo":137,"name":138,"nameZh":25,"description":139,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":140,"categories":141,"keywords":142,"publishedAt":33},"01M7DEJDM4Z8MVVJPAFFRSQMK3","szl-repo-pin","Records one SHA-256 pin over every Git repository an analysis depends on: HEAD commit, clean or dirty tree, exact tag, configured origin, for each declared repository, and a composite digest that exists only when every tree is clean; later re-reads them and reports MATCH \u002F DRIFT \u002F DIRTY \u002F MISSING per repository. Use when an analysis spans several repositories (analysis code, data preparation, a vendored utility library), when a capsule or session receipt needs to name the code versions, or when a collaborator asks 'which commits did you actually run'. Not a statement that the commits are published or correct.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-repo-pin\u002FSKILL.md",[],[],{"skillNo":144,"name":145,"nameZh":25,"description":146,"descriptionZh":25,"repositoryFullName":27,"ownerLogin":5,"ownerAvatarUrl":7,"ownerAccountNo":4,"ownerDisplayName":6,"repositoryNo":28,"starsCount":13,"licenseSpdxId":29,"licenseFileUrl":25,"sourceHtmlUrl":147,"categories":148,"keywords":149,"publishedAt":33},"01M7DEJ6T0K9ESVBD7YGQ9YYG2","szl-release-continuity","Trace supplied immutable release observations across GitHub, PyPI, Hugging Face models, datasets and Spaces, and a runtime. Compare declared revisions, canonical source identities and per-surface artifact hashes; retain missing evidence, conflicts and refused readiness. Use when a green CI run, published wheel, Hub listing or running Space is being treated as an end-to-end release. Offline, bounded, standard library only; no credentials.","https:\u002F\u002Fgithub.com\u002Fszl-holdings\u002Fszl-skills\u002Fblob\u002Fmain\u002Fskills\u002Fszl-release-continuity\u002FSKILL.md",[],[],18]