[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7FCDYWK1X6YGDK370KXYF2X":3,"skill-same:01M7FCDYWK1X6YGDK370KXYF2X":224},{"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":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7FCDYWK1X6YGDK370KXYF2X","thumbnail-skill",null,"Deterministic thumbnail rendering execution Skill for the AI Video Production Ecosystem. Use it when a caller (normally video-production-agent) has already decided exactly what a thumbnail should show - which still image or which explicit video timestamp, what text, what layout - and needs it rendered safely to PNG\u002FJPEG with provenance. Do NOT use it to choose what a thumbnail should show, pick a \"best\" video frame, detect faces, generate a title, judge design quality, or run arbitrary ffmpeg\u002Fimage commands (it refuses them); those decisions belong to the caller.","kajisho5\u002Fskillpact","kajisho5","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F20648875?v=4","01M7F3FS60ADGJ2CM2PW6RDFR2","Shogo Kajiyama","01M7FC4SG7BXWJK655CZ2Q42J8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: thumbnail-skill\ndescription: Deterministic thumbnail rendering execution Skill for the AI Video Production Ecosystem. Use it when a caller (normally video-production-agent) has already decided exactly what a thumbnail should show - which still image or which explicit video timestamp, what text, what layout - and needs it rendered safely to PNG\u002FJPEG with provenance. Do NOT use it to choose what a thumbnail should show, pick a \"best\" video frame, detect faces, generate a title, judge design quality, or run arbitrary ffmpeg\u002Fimage commands (it refuses them); those decisions belong to the caller.\n---\n\n# thumbnail-skill\n\nMachine interface: `thumbnail run - --json` with a `{\"tool\": \"thumbnail\u002F\u003Cname>\", \"params\": {...}}`\ndocument on stdin; exactly one response document on stdout. `thumbnail skill --json` prints the\ncontract, `thumbnail doctor --json` the environment (fonts, ffmpeg-skill, path policy).\n\nTools: `thumbnail\u002Fvalidate` (structural check only), `thumbnail\u002Frender` (canvas + assets + elements\n-> PNG\u002FJPEG), `thumbnail\u002Fextract_frame` (one video timestamp -> one frame, nothing else).\n\nRules for a calling agent:\n1. Decide everything about content before calling this skill: which still image, or which exact\n   video timestamp (`asset.timestamp`, seconds); what the text says (with literal `\\n` for line\n   breaks — this skill never wraps or reflows text); where every element sits (`position`, `size`,\n   `z_index`). This skill renders exactly what it is given and searches for nothing.\n2. Reference fonts by `font_id` from the registry (`thumbnail skill --json` -> `fonts.font_ids`, or\n   `thumbnail doctor --json` for what actually resolves on this machine). A request never carries a\n   font path; an unresolvable `font_id` fails `MISSING_INPUT`, never a silent substitution.\n3. Never send `command`, `argv`, `shell`, `filter`\u002F`filter_complex`\u002F`vf`\u002F`af`, `executable`, `env`,\n   or HTML\u002FCSS\u002FJS: the request is rejected with `INVALID_REQUEST` wherever such a key appears.\n4. Keep outputs inside the workspace (`--workspace` \u002F `options.workspace`), never at an input path;\n   set `overwrite: true` deliberately. Restrict inputs with `options.allowed_input_roots` when the\n   caller's own inputs come from an untrusted or shared location.\n5. Read `output`, `sha256`, `width`\u002F`height`, `reused` and `provenance` from the response; the exit\n   code is 0 only when `ok` is true. A `video_frame` asset needs ffmpeg-skill reachable (env var\n   `THUMBNAIL_SKILL_FFMPEG_SKILL_DIR` or `--ffmpeg-skill`); a still-image-only document never needs it.\n\nSee README.md for the ThumbnailDocument schema, element\u002Ffont\u002Foutput tables, error codes, security\nboundary and the responsibility split with video-production-agent and the other media skills.\n","# thumbnail-skill\n\nMachine interface: `thumbnail run - --json` with a `{\"tool\": \"thumbnail\u002F\u003Cname>\", \"params\": {...}}`\ndocument on stdin; exactly one response document on stdout. `thumbnail skill --json` prints the\ncontract, `thumbnail doctor --json` the environment (fonts, ffmpeg-skill, path policy).\n\nTools: `thumbnail\u002Fvalidate` (structural check only), `thumbnail\u002Frender` (canvas + assets + elements\n-> PNG\u002FJPEG), `thumbnail\u002Fextract_frame` (one video timestamp -> one frame, nothing else).\n\nRules for a calling agent:\n1. Decide everything about content before calling this skill: which still image, or which exact\n   video timestamp (`asset.timestamp`, seconds); what the text says (with literal `\\n` for line\n   breaks — this skill never wraps or reflows text); where every element sits (`position`, `size`,\n   `z_index`). This skill renders exactly what it is given and searches for nothing.\n2. Reference fonts by `font_id` from the registry (`thumbnail skill --json` -> `fonts.font_ids`, or\n   `thumbnail doctor --json` for what actually resolves on this machine). A request never carries a\n   font path; an unresolvable `font_id` fails `MISSING_INPUT`, never a silent substitution.\n3. Never send `command`, `argv`, `shell`, `filter`\u002F`filter_complex`\u002F`vf`\u002F`af`, `executable`, `env`,\n   or HTML\u002FCSS\u002FJS: the request is rejected with `INVALID_REQUEST` wherever such a key appears.\n4. Keep outputs inside the workspace (`--workspace` \u002F `options.workspace`), never at an input path;\n   set `overwrite: true` deliberately. Restrict inputs with `options.allowed_input_roots` when the\n   caller's own inputs come from an untrusted or shared location.\n5. Read `output`, `sha256`, `width`\u002F`height`, `reused` and `provenance` from the response; the exit\n   code is 0 only when `ok` is true. A `video_frame` asset needs ffmpeg-skill reachable (env var\n   `THUMBNAIL_SKILL_FFMPEG_SKILL_DIR` or `--ffmpeg-skill`); a still-image-only document never needs it.\n\nSee README.md for the ThumbnailDocument schema, element\u002Ffont\u002Foutput tables, error codes, security\nboundary and the responsibility split with video-production-agent and the other media skills.\n",{"name":5,"description":7},"main","962dae6e9e2cd9a6c705fe18af94cd8689ac1a87","samples\u002Fthumbnail-skill","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact","https:\u002F\u002Fgithub.com\u002Fkajisho5","https:\u002F\u002Fseventh-well.com\u002F","Video \u002F Audio \u002F Streaming Engineer\r\nBuilding tools for real-world event production.\r\nAI × FFmpeg × Automation","2026-11-13 07:11:38",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add kajisho5\u002Fskillpact --skill thumbnail-skill",{"tool":39,"command":40},"bunx","bunx skills add kajisho5\u002Fskillpact --skill thumbnail-skill",{"tool":42,"command":43},"pnpm","pnpm dlx skills add kajisho5\u002Fskillpact --skill thumbnail-skill","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002FSKILL.md",true,[47,53,57,63,69,73,77,80,84,88,92,96,100,104,108,112,116,120,124,128,132,138,142,146,150,154,158,162,166,170,174,178,182,186,189,193,196,200,204,208,212,216,220],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},".github\u002FFUNDING.yml","REFERENCE","参考文档",17,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002F.github\u002FFUNDING.yml",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},".github\u002Fworkflows\u002Ftests.yml",2957,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002F.github\u002Fworkflows\u002Ftests.yml",{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},".gitignore","OTHER","其他",98,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002F.gitignore",{"path":64,"resourceType":65,"resourceTypeDesc":66,"sizeBytes":67,"sourceHtmlUrl":68},"brand\u002Fapp-icon-dark.png","ASSET","资产文件",57813,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Fapp-icon-dark.png",{"path":70,"resourceType":65,"resourceTypeDesc":66,"sizeBytes":71,"sourceHtmlUrl":72},"brand\u002Fapp-icon.png",57807,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Fapp-icon.png",{"path":74,"resourceType":65,"resourceTypeDesc":66,"sizeBytes":75,"sourceHtmlUrl":76},"brand\u002Flogo\u002Fthumbnail-skill-wordmark-black.svg",4063,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Flogo\u002Fthumbnail-skill-wordmark-black.svg",{"path":78,"resourceType":65,"resourceTypeDesc":66,"sizeBytes":75,"sourceHtmlUrl":79},"brand\u002Flogo\u002Fthumbnail-skill-wordmark-white.svg","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Flogo\u002Fthumbnail-skill-wordmark-white.svg",{"path":81,"resourceType":65,"resourceTypeDesc":66,"sizeBytes":82,"sourceHtmlUrl":83},"brand\u002Fthumbnail-skill.icon\u002FAssets\u002Fframe.svg",319,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Fthumbnail-skill.icon\u002FAssets\u002Fframe.svg",{"path":85,"resourceType":65,"resourceTypeDesc":66,"sizeBytes":86,"sourceHtmlUrl":87},"brand\u002Fthumbnail-skill.icon\u002FAssets\u002Fscene.svg",414,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Fthumbnail-skill.icon\u002FAssets\u002Fscene.svg",{"path":89,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":90,"sourceHtmlUrl":91},"brand\u002Fthumbnail-skill.icon\u002Ficon.json",1745,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fbrand\u002Fthumbnail-skill.icon\u002Ficon.json",{"path":93,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":94,"sourceHtmlUrl":95},"docs\u002Farchitecture.md",5339,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fdocs\u002Farchitecture.md",{"path":97,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":98,"sourceHtmlUrl":99},"docs\u002Fdecisions.md",5963,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fdocs\u002Fdecisions.md",{"path":101,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":102,"sourceHtmlUrl":103},"docs\u002Fffmpeg-skill.md",6762,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fdocs\u002Fffmpeg-skill.md",{"path":105,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":106,"sourceHtmlUrl":107},"docs\u002Fmodel.md",4926,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fdocs\u002Fmodel.md",{"path":109,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":110,"sourceHtmlUrl":111},"docs\u002Fsecurity.md",5557,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fdocs\u002Fsecurity.md",{"path":113,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":114,"sourceHtmlUrl":115},"docs\u002Ftesting.md",3774,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fdocs\u002Ftesting.md",{"path":117,"resourceType":117,"resourceTypeDesc":118,"sizeBytes":119,"sourceHtmlUrl":16},"LICENSE","许可证",1065,{"path":121,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":122,"sourceHtmlUrl":123},"pyproject.toml",868,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fpyproject.toml",{"path":125,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":126,"sourceHtmlUrl":127},"README.md",17900,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002FREADME.md",{"path":129,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":130,"sourceHtmlUrl":131},"skill.contract.json",9853,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fskill.contract.json",{"path":133,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":136,"sourceHtmlUrl":137},"src\u002Fthumbnail_skill\u002F__init__.py","SCRIPT","脚本",1068,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002F__init__.py",{"path":139,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":140,"sourceHtmlUrl":141},"src\u002Fthumbnail_skill\u002Fadapter.py",21604,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fadapter.py",{"path":143,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":144,"sourceHtmlUrl":145},"src\u002Fthumbnail_skill\u002Fcanonical.py",917,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fcanonical.py",{"path":147,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":148,"sourceHtmlUrl":149},"src\u002Fthumbnail_skill\u002Fcli.py",10992,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fcli.py",{"path":151,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":152,"sourceHtmlUrl":153},"src\u002Fthumbnail_skill\u002Fcontract.py",10770,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fcontract.py",{"path":155,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":156,"sourceHtmlUrl":157},"src\u002Fthumbnail_skill\u002Fdoctor.py",3992,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fdoctor.py",{"path":159,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":160,"sourceHtmlUrl":161},"src\u002Fthumbnail_skill\u002Ferrors.py",2851,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Ferrors.py",{"path":163,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":164,"sourceHtmlUrl":165},"src\u002Fthumbnail_skill\u002Fexecutor.py",25277,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fexecutor.py",{"path":167,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":168,"sourceHtmlUrl":169},"src\u002Fthumbnail_skill\u002Ffonts.py",6584,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Ffonts.py",{"path":171,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":172,"sourceHtmlUrl":173},"src\u002Fthumbnail_skill\u002Fmodel.py",27546,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fmodel.py",{"path":175,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":176,"sourceHtmlUrl":177},"src\u002Fthumbnail_skill\u002Frenderer.py",7890,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Frenderer.py",{"path":179,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":180,"sourceHtmlUrl":181},"src\u002Fthumbnail_skill\u002Fsecurity.py",8149,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fsecurity.py",{"path":183,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":184,"sourceHtmlUrl":185},"src\u002Fthumbnail_skill\u002Fskill.py",2392,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Fsrc\u002Fthumbnail_skill\u002Fskill.py",{"path":187,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":14,"sourceHtmlUrl":188},"tests\u002F__init__.py","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002F__init__.py",{"path":190,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":191,"sourceHtmlUrl":192},"tests\u002Fconftest.py",1947,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Fconftest.py",{"path":194,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":14,"sourceHtmlUrl":195},"tests\u002Ffixtures\u002F__init__.py","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ffixtures\u002F__init__.py",{"path":197,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":198,"sourceHtmlUrl":199},"tests\u002Ffixtures\u002Fgenerate.py",2483,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ffixtures\u002Fgenerate.py",{"path":201,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":202,"sourceHtmlUrl":203},"tests\u002Ftest_cli.py",6997,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ftest_cli.py",{"path":205,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":206,"sourceHtmlUrl":207},"tests\u002Ftest_contract.py",2305,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ftest_contract.py",{"path":209,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":210,"sourceHtmlUrl":211},"tests\u002Ftest_ffmpeg_skill_compat.py",9572,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ftest_ffmpeg_skill_compat.py",{"path":213,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":214,"sourceHtmlUrl":215},"tests\u002Ftest_integration.py",22060,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ftest_integration.py",{"path":217,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":218,"sourceHtmlUrl":219},"tests\u002Ftest_security.py",8022,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ftest_security.py",{"path":221,"resourceType":134,"resourceTypeDesc":135,"sizeBytes":222,"sourceHtmlUrl":223},"tests\u002Ftest_unit.py",9260,"https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fthumbnail-skill\u002Ftests\u002Ftest_unit.py",[225,233,241,249],{"skillNo":226,"name":227,"nameZh":6,"description":228,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":229,"sourceHtmlUrl":230,"categories":231,"keywords":232,"publishedAt":20},"01M7FCE4B3HJXKX3782K4Q5Y0M","video-editing-skill","Deterministic video editing from a typed edit request (trim, cut, concat with transitions, speed change, fit \u002F fill \u002F resize, still-image overlay) with an exact source-to-timeline mapping and per-operation provenance. Executes through ffmpeg-skill; never takes commands, argv, filter strings or executables. Use when an agent has already decided what to edit and needs the edit performed, verified and traced. Not for deciding what to cut, transcribing, captioning, colour grading or audio mastering.","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fvideo-editing-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fvideo-editing-skill\u002FSKILL.md",[],[],{"skillNo":234,"name":235,"nameZh":6,"description":236,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":237,"sourceHtmlUrl":238,"categories":239,"keywords":240,"publishedAt":20},"01M7FCDHRY5H7A8B8A72CSDZ5S","subtitle-skill","Deterministic subtitle validation, SRT\u002FWebVTT generation, and video caption burn-in. Given an already-decided, typed subtitle document (cue text, timing, optional speaker\u002Fstyle), validate its timeline, generate an SRT or WebVTT file, or burn it into a video by delegating to ffmpeg-skill's caption tool. Use this skill only after another agent has already decided what the subtitle cues should say and how they should be split -- it performs no transcription, no diarization, no cue-splitting judgement, and no \"does this look right\" visual check.","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fsubtitle-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fsubtitle-skill\u002FSKILL.md",[],[],{"skillNo":242,"name":243,"nameZh":6,"description":244,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":245,"sourceHtmlUrl":246,"categories":247,"keywords":248,"publishedAt":20},"01M7FCD6SWSSZ330YP6X6Z4TD9","ffmpeg-skill","Edit video and audio with local FFmpeg from natural-language requests: cut, trim, join, resize\u002Freframe (9:16, 1:1), speed change, captions and subtitles (SRT\u002FASS, animated, karaoke), logos and text overlays, lower-thirds and titles, silence removal, multicam and external-mic sync, loudness normalisation, HDR\u002FDolby Vision to SDR, LUTs, background music with ducking, platform exports (YouTube, Reels, TikTok, X), compliance checks, scene detection and highlight reels, contact sheets to inspect results, and whole-edit project files. Use this skill whenever the user mentions a video or audio file (mp4, mov, mkv, wav, m4a), footage, a clip, captions, subtitles, a reel or short, YouTube\u002FInstagram\u002FTikTok delivery, LUFS, sync, transcoding, ffmpeg, or asks to make something \"60 seconds\", \"vertical\", \"louder\", \"captioned\" — even when they do not say \"edit\". Python 3.9 standard library only, no cloud, no API keys.","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fffmpeg-skill\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002Fsamples\u002Fffmpeg-skill\u002FSKILL.md",[],[],{"skillNo":250,"name":251,"nameZh":6,"description":252,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":253,"licenseFileUrl":254,"sourceHtmlUrl":255,"categories":256,"keywords":257,"publishedAt":20},"01M7FC4XRQQDVAKX4V023SEQP5","skillpact","Write and check SkillPact contracts (skill.contract.json) for Agent Skills with the skillpact CLI - draft a contract from a SKILL.md, validate it, run its conformance tests network-isolated, hash and sign every file, report the level L1 \u002F L2 \u002F L3, match it against a harness policy (allow \u002F ask \u002F deny) and compile that decision to Claude Code, Cursor, Codex or Gemini CLI settings or export it to HushSpec, Rego, APS or ACS. Use when the user wants a skill's permissions declared or reviewed before install, a contract drafted, validated, signed or migrated, a conformance level or badge, a policy check, or harness permission settings made from a policy. Python 3.11+; bubblewrap on Linux for L2 and L3. Not a runtime sandbox and not a malware scanner - L3 means these files, signed by a trusted author, passed their tests, not that the skill is safe.","CC-BY-4.0","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fkajisho5\u002Fskillpact\u002Fblob\u002Fmain\u002FSKILL.md",[],[]]