[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71AXNZH6D319JYHB2RC95NP":3,"skill-same:01M71AXNZH6D319JYHB2RC95NP":86},{"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":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":6,"ownerHtmlUrl":52,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":53,"install":54,"resources":67},"01M71AXNZH6D319JYHB2RC95NP","libfuzzer-fuzzing-current",null,"Set up, review or debug bounded C\u002FC++ libFuzzer harnesses, sanitizer builds, corpus replay and crash triage. Use when qualifying LLVMFuzzerTestOneInput, diagnosing missing findings or coverage, or choosing reproducible campaign limits and instrumentation.","strmt7\u002Fai_skills_collection_benchmarked","strmt7","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F28989754?v=4","01M71929P914SWKQ7QJV24NBBY","Stratos","01M71929Q9VGKQ7PWYR95VSQQQ",1,"CC-BY-SA-4.0","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current\u002FSKILL.md",[],[20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"libfuzzer fuzzing current","libfuzzer","fuzzing","replay","when","limits","sanitizer","harnesses","missing","c++","instrumentation","campaign","up","crash","use","coverage","SET","Reproducible","debug","triage","2026-11-10 10:12:30",false,"---\nname: libfuzzer-fuzzing-current\ndescription: Set up, review or debug bounded C\u002FC++ libFuzzer harnesses, sanitizer builds, corpus replay and crash triage. Use when qualifying LLVMFuzzerTestOneInput, diagnosing missing findings or coverage, or choosing reproducible campaign limits and instrumentation.\nlicense: CC-BY-SA-4.0\nmetadata:\n  status: experimental-unbenchmarked\n---\n\n# Qualified C\u002FC++ fuzzing\n\nStart with the actual target's format, valid byte domain, documented rejection\nbehavior and independently defined wrong outcomes. A harness must exercise the\ntarget and make meaningful results observable. Keep harness faults distinct from\ntarget defects. Include empty and malformed inputs when they belong to the API;\ndo not reject them merely to silence a crash. Reset state between callbacks and\njoin target threads before returning.\n\nDecode integer fields with a specified byte order or a qualified provider.\nAvoid casts that assume alignment and object access. Bound provider fields against\nthe target's contract, including length, terminator and allocation ownership.\n`ConsumeBytesWithTerminator` returns at least its terminator; determine whether\nthat byte belongs in the target's length. Do not call `free` on an arbitrary\nlibrary-owned or differently allocated result.\n\nVerify a supported stable compiler and its matching libFuzzer\u002Fsanitizer runtimes\nbefore installing. Record artifact digests and platform. As of 2026-10-07,\nLLVM 23.1.3 is the observed stable release; Linux x64 native known-input controls\nare qualified here. Other platforms and dependencies need their own checks.\nlibFuzzer receives important bug fixes; it does not promise major new features.\n\nInstrument target libraries with `-fsanitize=fuzzer-no-link` and the selected\nsanitizers; link the final driver with `-fsanitize=fuzzer` and the same toolchain.\nKeep target, driver, dependency and sanitizer flags explicit. Do not assume a\nmacro removes an existing `main`, or mix incompatible compiler runtimes. Verify\nthe complete build and replay a known instrumented defect plus a valid input.\nAn unused expression may disappear under optimization; ASan alone does not\ndetect every undefined arithmetic operation. Preserve production hardening\nunless a specific, recorded incompatibility justifies an experiment.\n\nUse an owned disposable environment with explicit wall-clock, per-input, RSS,\nallocation, process, input-size, corpus, artifact and output limits. ASan's shadow\naddress reservation is not RSS. `-rss_limit_mb` bounds resident memory; disabling\nit does not repair address-space limits. Check the actual runtime's options.\nDo not disable leak detection or continue after memory corruption to manufacture\nclean results; record any qualified exclusion separately.\n\nStart with a bounded single-process smoke, then a measured campaign. Choose one\nqualified parallel mode and its total resource budget: jobs count campaigns,\nworkers bound their concurrency, and fork mode adds child processes. Keep crash,\ntimeout and out-of-memory artifacts distinguishable. A graceful bounded exit is\ndifferent from a timeout that kills collection. Record complete raw outcomes.\n\nPin source, binary, corpus, dictionary and environment identities. Validate\ndictionary syntax rather than passing unescaped extracted strings to the parser.\nFeature-preserving corpus minimization is not crash-input minimization or proof\nof semantic coverage. Keep original crash seeds, minimize copies, and replay both\nwith the same binary and instrumentation. A fixed seed alone does not reproduce\nscheduling, mutable state or corpus changes.\n\nUse matching LLVM source-based tools for source coverage and report its scope,\ndenominators, failures and profile compatibility separately from libFuzzer's\nfeedback counters. Verify coverage collection across normal and crashing exits.\nChoose another engine only after qualifying its harness, build and reset\ncontracts. Do not infer efficacy, bug-discovery speed or agent superiority from\nan installation check, known-seed replay or one favorable run.\n\nSee [qualified scope](references\u002Fqualification.md). Original instructions and\nattribution are preserved in this overlay; coding-agent efficacy is unbenchmarked.\n","# Qualified C\u002FC++ fuzzing\n\nStart with the actual target's format, valid byte domain, documented rejection\nbehavior and independently defined wrong outcomes. A harness must exercise the\ntarget and make meaningful results observable. Keep harness faults distinct from\ntarget defects. Include empty and malformed inputs when they belong to the API;\ndo not reject them merely to silence a crash. Reset state between callbacks and\njoin target threads before returning.\n\nDecode integer fields with a specified byte order or a qualified provider.\nAvoid casts that assume alignment and object access. Bound provider fields against\nthe target's contract, including length, terminator and allocation ownership.\n`ConsumeBytesWithTerminator` returns at least its terminator; determine whether\nthat byte belongs in the target's length. Do not call `free` on an arbitrary\nlibrary-owned or differently allocated result.\n\nVerify a supported stable compiler and its matching libFuzzer\u002Fsanitizer runtimes\nbefore installing. Record artifact digests and platform. As of 2026-10-07,\nLLVM 23.1.3 is the observed stable release; Linux x64 native known-input controls\nare qualified here. Other platforms and dependencies need their own checks.\nlibFuzzer receives important bug fixes; it does not promise major new features.\n\nInstrument target libraries with `-fsanitize=fuzzer-no-link` and the selected\nsanitizers; link the final driver with `-fsanitize=fuzzer` and the same toolchain.\nKeep target, driver, dependency and sanitizer flags explicit. Do not assume a\nmacro removes an existing `main`, or mix incompatible compiler runtimes. Verify\nthe complete build and replay a known instrumented defect plus a valid input.\nAn unused expression may disappear under optimization; ASan alone does not\ndetect every undefined arithmetic operation. Preserve production hardening\nunless a specific, recorded incompatibility justifies an experiment.\n\nUse an owned disposable environment with explicit wall-clock, per-input, RSS,\nallocation, process, input-size, corpus, artifact and output limits. ASan's shadow\naddress reservation is not RSS. `-rss_limit_mb` bounds resident memory; disabling\nit does not repair address-space limits. Check the actual runtime's options.\nDo not disable leak detection or continue after memory corruption to manufacture\nclean results; record any qualified exclusion separately.\n\nStart with a bounded single-process smoke, then a measured campaign. Choose one\nqualified parallel mode and its total resource budget: jobs count campaigns,\nworkers bound their concurrency, and fork mode adds child processes. Keep crash,\ntimeout and out-of-memory artifacts distinguishable. A graceful bounded exit is\ndifferent from a timeout that kills collection. Record complete raw outcomes.\n\nPin source, binary, corpus, dictionary and environment identities. Validate\ndictionary syntax rather than passing unescaped extracted strings to the parser.\nFeature-preserving corpus minimization is not crash-input minimization or proof\nof semantic coverage. Keep original crash seeds, minimize copies, and replay both\nwith the same binary and instrumentation. A fixed seed alone does not reproduce\nscheduling, mutable state or corpus changes.\n\nUse matching LLVM source-based tools for source coverage and report its scope,\ndenominators, failures and profile compatibility separately from libFuzzer's\nfeedback counters. Verify coverage collection across normal and crashing exits.\nChoose another engine only after qualifying its harness, build and reset\ncontracts. Do not infer efficacy, bug-discovery speed or agent superiority from\nan installation check, known-seed replay or one favorable run.\n\nSee [qualified scope](references\u002Fqualification.md). Original instructions and\nattribution are preserved in this overlay; coding-agent efficacy is unbenchmarked.\n",{"name":5,"license":15,"metadata":45,"description":7},{"status":46},"experimental-unbenchmarked","main","5c95bcac4646c5aed06bd9aed5235cfc7985aafc","included\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current",0,"https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked","https:\u002F\u002Fgithub.com\u002Fstrmt7","2026-11-11 04:18:12",{"source":8,"skillName":5,"commands":55,"prompt":65,"zipAvailable":66},[56,59,62],{"tool":57,"command":58},"npx","npx skills add strmt7\u002Fai_skills_collection_benchmarked --skill libfuzzer-fuzzing-current",{"tool":60,"command":61},"bunx","bunx skills add strmt7\u002Fai_skills_collection_benchmarked --skill libfuzzer-fuzzing-current",{"tool":63,"command":64},"pnpm","pnpm dlx skills add strmt7\u002Fai_skills_collection_benchmarked --skill libfuzzer-fuzzing-current","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current\u002FSKILL.md",true,[68,72,78,82],{"path":69,"resourceType":69,"resourceTypeDesc":70,"sizeBytes":71,"sourceHtmlUrl":16},"LICENSE","许可证",20131,{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"NOTICE.md","REFERENCE","参考文档",768,"https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current\u002FNOTICE.md",{"path":79,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":80,"sourceHtmlUrl":81},"provenance\u002Foriginal-SKILL.md",24720,"https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current\u002Fprovenance\u002Foriginal-SKILL.md",{"path":83,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Fqualification.md",1809,"https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fimproved\u002Fskills\u002Flibfuzzer-fuzzing-current\u002Freferences\u002Fqualification.md",[87,121,150,176,201,229],{"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":91,"licenseFileUrl":92,"sourceHtmlUrl":93,"categories":94,"keywords":103,"publishedAt":40},"01M71D4H3T6M57DFKN8HENAWRV","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Ftesting-qa-benchmarking\u002Fofficial-reference\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Ftesting-qa-benchmarking\u002Fofficial-reference\u002Fwebapp-testing\u002FSKILL.md",[95,100],{"categoryNo":96,"categoryName":97,"taxonomyType":98,"taxonomyTypeDesc":99},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":101,"categoryName":102,"taxonomyType":98,"taxonomyTypeDesc":99},"01JBSEEDCAT000000000000009","前端开发",[104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120],"webapp","testing","webapp testing","for","applications","local","frontend","debugging","toolkit","logs","browser","with","playwright","web","behavior","ui","screenshots",{"skillNo":122,"name":123,"nameZh":6,"description":124,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":91,"licenseFileUrl":125,"sourceHtmlUrl":126,"categories":127,"keywords":131,"publishedAt":40},"01M71D4CQR2A2TB7GASCR5ZBG3","skill-creator","Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Ftesting-qa-benchmarking\u002Fofficial-reference\u002Fskill-creator-anthropics\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Ftesting-qa-benchmarking\u002Fofficial-reference\u002Fskill-creator-anthropics\u002FSKILL.md",[128],{"categoryNo":129,"categoryName":130,"taxonomyType":98,"taxonomyTypeDesc":99},"01JBSEEDCAT000000000000023","Skill与智能体开发",[132,133,134,135,136,137,24,138,139,140,141,142,143,144,145,146,147,148,107,149],"skill creator","creator","performance","improve","run evals","new","analysis","create","optimize","variance analysis","run","edit","benchmark","a skill","existing skill","evals","modify","accuracy",{"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":154,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":159,"publishedAt":40},"01M71D488XHEZVXXSXFYP8ENMH","playwright-skill","Complete browser automation with Playwright. Auto-detects dev servers, writes reusable test scripts, and supports screenshots, responsive checks, UX validation, login flows, link checks, and arbitrary browser automation. Use when the user wants to test a website, automate browser interactions, validate web functionality, or perform browser-based testing.","MIT","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Ftesting-qa-benchmarking\u002Flatest-release-browser-automation\u002Fplaywright-skill\u002FSKILL.md",[157,158],{"categoryNo":96,"categoryName":97,"taxonomyType":98,"taxonomyTypeDesc":99},{"categoryNo":129,"categoryName":130,"taxonomyType":98,"taxonomyTypeDesc":99},[116,160,161,162,105,163,164,24,165,114,166,167,168,169,170,171,172,173,174,175],"playwright skill","browser automation","flows","auto","access","Bash","website","ux","login","scripts","read","write","node","install","network","responsive",{"skillNo":177,"name":178,"nameZh":6,"description":179,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":91,"licenseFileUrl":180,"sourceHtmlUrl":181,"categories":182,"keywords":184,"publishedAt":40},"01M71CY25GM939M6RES91QSG3T","web-artifacts-builder","Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn\u002Fui). Use for complex artifacts requiring state management, routing, or shadcn\u002Fui components - not for simple single-file HTML\u002FJSX artifacts.","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Fsearch-retrieval-web-automation\u002Fofficial-reference\u002Fweb-artifacts-builder\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Fsearch-retrieval-web-automation\u002Fofficial-reference\u002Fweb-artifacts-builder\u002FSKILL.md",[183],{"categoryNo":101,"categoryName":102,"taxonomyType":98,"taxonomyTypeDesc":99},[185,186,187,117,188,189,190,107,191,192,110,193,34,194,195,196,197,198,199,200],"builder","artifacts builder","artifacts","web artifacts builder","single-file","css","Tailwind","ui components","single","file","components","react","multi","state","routing","state management",{"skillNo":202,"name":203,"nameZh":6,"description":204,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":91,"licenseFileUrl":205,"sourceHtmlUrl":206,"categories":207,"keywords":217,"publishedAt":40},"01M71CXZB60QJJ1H4S3NN94WMZ","frontend-design","Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML\u002FCSS layouts, or when styling\u002Fbeautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Fsearch-retrieval-web-automation\u002Fofficial-reference\u002Ffrontend-design\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Fsearch-retrieval-web-automation\u002Fofficial-reference\u002Ffrontend-design\u002FSKILL.md",[208,209,212],{"categoryNo":101,"categoryName":102,"taxonomyType":98,"taxonomyTypeDesc":99},{"categoryNo":210,"categoryName":211,"taxonomyType":98,"taxonomyTypeDesc":99},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":213,"categoryName":214,"taxonomyType":215,"taxonomyTypeDesc":216},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[110,218,219,190,220,221,222,108,223,224,34,24,187,225,195,226,196,227,228,115],"design","frontend design","ui design","generic","aesthetics","landing","pages","code","web ui","react components","examples",{"skillNo":230,"name":231,"nameZh":6,"description":232,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":154,"licenseFileUrl":6,"sourceHtmlUrl":233,"categories":234,"keywords":235,"publishedAt":40},"01M71CVZ54WFBTVA7R16PFY1ZE","zarr-python","Chunked N-D arrays for cloud storage. Compressed arrays, parallel I\u002FO, S3\u002FGCS integration, NumPy\u002FDask\u002FXarray compatible, for large-scale scientific computing pipelines.","https:\u002F\u002Fgithub.com\u002Fstrmt7\u002Fai_skills_collection_benchmarked\u002Fblob\u002Fmain\u002Fincluded\u002Fskills\u002Fby-category\u002Fscience-research-data-analysis\u002Flatest-release-community\u002Fzarr-python\u002FSKILL.md",[],[236,237,238,239,240,241,242,243,244,107,245,246,247,248,249,250,251,252],"zarr python","python","zarr","arrays","Cloud Storage","scientific","storage","parallel","dask","numpy","gcs","s3","integration","scientific computing","cloud","scale","pipelines"]