[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DE9XVZASJS9PC1A5BVMVHG":3,"skill-same:01M7DE9XVZASJS9PC1A5BVMVHG":132},{"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":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7DE9XVZASJS9PC1A5BVMVHG","verify-video-archive",null,"Use when validating the macOS or Linux archive workflow reached by `just convert-video`, including real media, prerequisites, terminal progress, source safety, locking, and recovery.","pascalandy\u002Fskills","pascalandy","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6694151?v=4","01M7DDRR1T4AMR6J3FV6RHHV5M","Pascal Andy","01M7DDRR2B04BJ8S7QZYZV1WFK",0,"MIT","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: \"verify-video-archive\"\ndescription: \"Use when validating the macOS or Linux archive workflow reached by `just convert-video`, including real media, prerequisites, terminal progress, source safety, locking, and recovery.\"\nkind: \"dev\"\n---\n\n# Verify video archive\n\nDrive the checkout implementation through `just convert-video`. The verifier\ncreates generated media and never reads the real archive or personal input\nfolders.\n\nLoad this skill from\n`authoring\u002Fverify-loops\u002Fverify-video-archive\u002FSKILL.md` in a skills repository\ncheckout. After skill installation, you can invoke\n`$verify-video-archive` and resolve `VERIFY_DIR` from the loaded skill directory.\n\n## Launch\n\nCreate a new run. The command answers `{\"ok\":true,\"file\":\"\u003Cmanifest>\"}`,\nwhere `file` is the retained manifest path.\n\n```bash\nVERIFY_DIR=\"\u003Cverify-video-archive-skill-dir>\"\nCHECKOUT=\"\u003Cdotfiles-checkout>\"\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  launch --checkout \"$CHECKOUT\"\n```\n\nCopy the `file` value into `MANIFEST`. Launch creates a unique\ndisposable root and a separate evidence root. Each scenario gets its own home,\ninput roots, archive, state, journal, and scratch directories.\n\n## Doctor\n\nRun Doctor before Drive.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  doctor --manifest \"$MANIFEST\"\n```\n\nDoctor checks macOS or Linux, Python 3.11 or newer, Just, Git, FFmpeg, FFprobe,\n`libx265`, a resolved `lsof` executable, and the checkout wrapper. An unmet\ncheck blocks real-media acceptance on that platform.\n\n## Drive\n\nRead [the feature map](features\u002FREADME.md), then run one feature or the full map.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  drive --manifest \"$MANIFEST\" --feature conversion\n\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  drive --manifest \"$MANIFEST\" --feature all\n```\n\nEvery archive invocation executes this argument array:\n\n```text\n\u003Cresolved-just> --justfile \u003Ccheckout>\u002Fjustfile convert-video\n```\n\nThe isolated `~\u002F.local\u002Fbin\u002Fvideo-archive` is a symbolic link to the checkout\nwrapper. It adds no archive behavior. The Just recipe supplies the real\n`archive` subcommand and uses the isolated default folders.\n\nThe recovery feature sets the verifier-only\n`VIDEO_ARCHIVE_VERIFY_PAUSE_AFTER_PREPARED_SECONDS` environment variable. The\narchive command pauses only after its durable `prepared` record. The verifier\nstops and kills the process group created for that invocation, then runs the\nsame Just command again without the pause.\n\nProgress scenarios run the real Just command in a task-owned PTY. The driver\nsets the PTY width with the terminal interface for the narrow-layout check. It\ndoes not fake outcome data. Generated media, archive state, exit status, and\nretained transcripts prove each outcome.\n\nThe dependency feature runs the real Just recipe with task-owned restricted\n`PATH` directories. It proves that missing Python, FFmpeg, FFprobe, `libx265`,\nor `lsof` fails before the source or archive changes.\n\nThe parallel-batch feature creates 100 generated videos for each serial,\nnormal, maximum, and cancellation run. It records selected limits, observed\nsimultaneous media processes, throughput, elapsed time, peak RSS, CPU use, and\nbackend-session applicability. During the cancellation run, it starts a second\nreal archive invocation and requires exit 75 before it sends one SIGINT to the\nfirst invocation. It then verifies exit 130, owned-child cleanup, recovery, and\nexact final accounting. A separate six-video 1280x720 workload exercises the\nmaximum-mode worker bound with representative media instead of relying on the\nsmall 100-video correctness fixtures for resource safety.\n\n## Evidence\n\nValidate the retained bundle after Drive.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  evidence --manifest \"$MANIFEST\"\n```\n\nRead `manifest.json` first. It records exact argument arrays, exit status,\ntranscripts, Git revision and dirty identity, platform and tool versions,\ninventories, hashes, FFprobe output, operation records, and check status.\n`skipped` and `unmet` remain separate from `passed`. `verifier` identifies\nthe B package and its files separately from `git` and `tested_files`, which\nidentify the A checkout under test.\n\nFor a target-platform acceptance run, use `--feature parallel-batch` and retain\nthe complete bundle. Run the feature on both Linux and Apple M2 hardware. A run\nfrom another machine does not satisfy either target gate.\n\nEvidence paths are relative to `manifest.json`. You can move the complete\nbundle and run Evidence again. It rechecks every retained path and SHA-256.\n\nThe verifier independently reopens each successful archive with FFprobe,\ndecodes it with FFmpeg, and compares its hash with the durable operation record.\nThe production store remains responsible for verifying publication before it\nremoves a source.\n\nSee [the evidence schema](references\u002Fevidence-schema.md) when reviewing or\nextending the manifest.\n\n## Cleanup\n\nRemove only the disposable root named by the manifest.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  cleanup --manifest \"$MANIFEST\"\n```\n\nCleanup checks the run ID and random owner token before removal. It refuses a\nsymbolic link, a broad path, a changed marker, or evidence stored below the run\nroot. Cleanup never loads a PID or signals a process. The evidence bundle and\nmanifest survive.\n\n## Helpers\n\nRun the whole lifecycle when you do not need to inspect the run before cleanup.\nIts answer names the manifest in `file`, also when a check fails.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  run --checkout \"$CHECKOUT\" --feature all\n```\n\n## Benchmark backends\n\nFor backend performance and compression qualification, read\n[the backend benchmark feature](features\u002Fbackend-benchmark.md). Its command\nuses generated media and invokes the same `just convert-video` entry point.\nKeep its visual gate unmet until a reviewer finishes the blinded comparison.\n\nRun focused helper tests after changing the verifier.\n\n```bash\nuv run --no-project python -m unittest discover \\\n  -s \"$VERIFY_DIR\u002Fscripts\u002Ftests\" -v\n```\n\nRun the verifier package checks in the skills checkout and the application\nchecks in the explicit dotfiles checkout. Each platform acceptance gate also\nneeds a passing full Drive run on that platform.\n\n```bash\njust check --only verify-video-archive\njust --justfile \"$CHECKOUT\u002Fjustfile\" test-video-archive\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  run --checkout \"$CHECKOUT\" --feature all\n```\n","# Verify video archive\n\nDrive the checkout implementation through `just convert-video`. The verifier\ncreates generated media and never reads the real archive or personal input\nfolders.\n\nLoad this skill from\n`authoring\u002Fverify-loops\u002Fverify-video-archive\u002FSKILL.md` in a skills repository\ncheckout. After skill installation, you can invoke\n`$verify-video-archive` and resolve `VERIFY_DIR` from the loaded skill directory.\n\n## Launch\n\nCreate a new run. The command answers `{\"ok\":true,\"file\":\"\u003Cmanifest>\"}`,\nwhere `file` is the retained manifest path.\n\n```bash\nVERIFY_DIR=\"\u003Cverify-video-archive-skill-dir>\"\nCHECKOUT=\"\u003Cdotfiles-checkout>\"\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  launch --checkout \"$CHECKOUT\"\n```\n\nCopy the `file` value into `MANIFEST`. Launch creates a unique\ndisposable root and a separate evidence root. Each scenario gets its own home,\ninput roots, archive, state, journal, and scratch directories.\n\n## Doctor\n\nRun Doctor before Drive.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  doctor --manifest \"$MANIFEST\"\n```\n\nDoctor checks macOS or Linux, Python 3.11 or newer, Just, Git, FFmpeg, FFprobe,\n`libx265`, a resolved `lsof` executable, and the checkout wrapper. An unmet\ncheck blocks real-media acceptance on that platform.\n\n## Drive\n\nRead [the feature map](features\u002FREADME.md), then run one feature or the full map.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  drive --manifest \"$MANIFEST\" --feature conversion\n\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  drive --manifest \"$MANIFEST\" --feature all\n```\n\nEvery archive invocation executes this argument array:\n\n```text\n\u003Cresolved-just> --justfile \u003Ccheckout>\u002Fjustfile convert-video\n```\n\nThe isolated `~\u002F.local\u002Fbin\u002Fvideo-archive` is a symbolic link to the checkout\nwrapper. It adds no archive behavior. The Just recipe supplies the real\n`archive` subcommand and uses the isolated default folders.\n\nThe recovery feature sets the verifier-only\n`VIDEO_ARCHIVE_VERIFY_PAUSE_AFTER_PREPARED_SECONDS` environment variable. The\narchive command pauses only after its durable `prepared` record. The verifier\nstops and kills the process group created for that invocation, then runs the\nsame Just command again without the pause.\n\nProgress scenarios run the real Just command in a task-owned PTY. The driver\nsets the PTY width with the terminal interface for the narrow-layout check. It\ndoes not fake outcome data. Generated media, archive state, exit status, and\nretained transcripts prove each outcome.\n\nThe dependency feature runs the real Just recipe with task-owned restricted\n`PATH` directories. It proves that missing Python, FFmpeg, FFprobe, `libx265`,\nor `lsof` fails before the source or archive changes.\n\nThe parallel-batch feature creates 100 generated videos for each serial,\nnormal, maximum, and cancellation run. It records selected limits, observed\nsimultaneous media processes, throughput, elapsed time, peak RSS, CPU use, and\nbackend-session applicability. During the cancellation run, it starts a second\nreal archive invocation and requires exit 75 before it sends one SIGINT to the\nfirst invocation. It then verifies exit 130, owned-child cleanup, recovery, and\nexact final accounting. A separate six-video 1280x720 workload exercises the\nmaximum-mode worker bound with representative media instead of relying on the\nsmall 100-video correctness fixtures for resource safety.\n\n## Evidence\n\nValidate the retained bundle after Drive.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  evidence --manifest \"$MANIFEST\"\n```\n\nRead `manifest.json` first. It records exact argument arrays, exit status,\ntranscripts, Git revision and dirty identity, platform and tool versions,\ninventories, hashes, FFprobe output, operation records, and check status.\n`skipped` and `unmet` remain separate from `passed`. `verifier` identifies\nthe B package and its files separately from `git` and `tested_files`, which\nidentify the A checkout under test.\n\nFor a target-platform acceptance run, use `--feature parallel-batch` and retain\nthe complete bundle. Run the feature on both Linux and Apple M2 hardware. A run\nfrom another machine does not satisfy either target gate.\n\nEvidence paths are relative to `manifest.json`. You can move the complete\nbundle and run Evidence again. It rechecks every retained path and SHA-256.\n\nThe verifier independently reopens each successful archive with FFprobe,\ndecodes it with FFmpeg, and compares its hash with the durable operation record.\nThe production store remains responsible for verifying publication before it\nremoves a source.\n\nSee [the evidence schema](references\u002Fevidence-schema.md) when reviewing or\nextending the manifest.\n\n## Cleanup\n\nRemove only the disposable root named by the manifest.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  cleanup --manifest \"$MANIFEST\"\n```\n\nCleanup checks the run ID and random owner token before removal. It refuses a\nsymbolic link, a broad path, a changed marker, or evidence stored below the run\nroot. Cleanup never loads a PID or signals a process. The evidence bundle and\nmanifest survive.\n\n## Helpers\n\nRun the whole lifecycle when you do not need to inspect the run before cleanup.\nIts answer names the manifest in `file`, also when a check fails.\n\n```bash\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  run --checkout \"$CHECKOUT\" --feature all\n```\n\n## Benchmark backends\n\nFor backend performance and compression qualification, read\n[the backend benchmark feature](features\u002Fbackend-benchmark.md). Its command\nuses generated media and invokes the same `just convert-video` entry point.\nKeep its visual gate unmet until a reviewer finishes the blinded comparison.\n\nRun focused helper tests after changing the verifier.\n\n```bash\nuv run --no-project python -m unittest discover \\\n  -s \"$VERIFY_DIR\u002Fscripts\u002Ftests\" -v\n```\n\nRun the verifier package checks in the skills checkout and the application\nchecks in the explicit dotfiles checkout. Each platform acceptance gate also\nneeds a passing full Drive run on that platform.\n\n```bash\njust check --only verify-video-archive\njust --justfile \"$CHECKOUT\u002Fjustfile\" test-video-archive\nuv run --no-project python \"$VERIFY_DIR\u002Fscripts\u002Fverify_video_archive.py\" \\\n  run --checkout \"$CHECKOUT\" --feature all\n```\n",{"kind":24,"name":5,"description":7},"dev","main","c2b929d1a54305e77aad0296872e90e08d68a637","skills\u002Fverify-video-archive",1,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills","https:\u002F\u002Fgithub.com\u002Fpascalandy","https:\u002F\u002Fblog.pascalandy.com","The question: how can I build the default one-stop place where entrepreneurs can start their mobile-first websites and blog? Psst psst… play-with-ghost.com","2026-11-15 03:34:59",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add pascalandy\u002Fskills --skill verify-video-archive",{"tool":40,"command":41},"bunx","bunx skills add pascalandy\u002Fskills --skill verify-video-archive",{"tool":43,"command":44},"pnpm","pnpm dlx skills add pascalandy\u002Fskills --skill verify-video-archive","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002FSKILL.md",true,[48,54,58,62,66,70,74,78,82,86,90,94,98,102,106,110,114,118,122,128],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"agents\u002Fopenai.yaml","REFERENCE","参考文档",271,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Fagents\u002Fopenai.yaml",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"features\u002Fbackend-benchmark.md",2001,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fbackend-benchmark.md",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"features\u002Fchanging-source.md",865,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fchanging-source.md",{"path":63,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":64,"sourceHtmlUrl":65},"features\u002Fcleanup-pending.md",1081,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fcleanup-pending.md",{"path":67,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":68,"sourceHtmlUrl":69},"features\u002Fcollisions.md",1050,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fcollisions.md",{"path":71,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":72,"sourceHtmlUrl":73},"features\u002Fconversion.md",1299,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fconversion.md",{"path":75,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":76,"sourceHtmlUrl":77},"features\u002Fdependencies.md",1001,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fdependencies.md",{"path":79,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":80,"sourceHtmlUrl":81},"features\u002Ffailure-continuation.md",1228,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Ffailure-continuation.md",{"path":83,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":84,"sourceHtmlUrl":85},"features\u002Finterruption.md",1033,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Finterruption.md",{"path":87,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":88,"sourceHtmlUrl":89},"features\u002Flock-contention.md",875,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Flock-contention.md",{"path":91,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":92,"sourceHtmlUrl":93},"features\u002Fopen-source.md",824,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fopen-source.md",{"path":95,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":96,"sourceHtmlUrl":97},"features\u002Fordering.md",855,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fordering.md",{"path":99,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":100,"sourceHtmlUrl":101},"features\u002Foriginal-fallback.md",1137,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Foriginal-fallback.md",{"path":103,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":104,"sourceHtmlUrl":105},"features\u002Foutput-contracts.md",1233,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Foutput-contracts.md",{"path":107,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":108,"sourceHtmlUrl":109},"features\u002Fparallel-batch.md",1939,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Fparallel-batch.md",{"path":111,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":112,"sourceHtmlUrl":113},"features\u002FREADME.md",3289,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002FREADME.md",{"path":115,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":116,"sourceHtmlUrl":117},"features\u002Frecovery.md",1525,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Ffeatures\u002Frecovery.md",{"path":119,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":120,"sourceHtmlUrl":121},"references\u002Fevidence-schema.md",4403,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Freferences\u002Fevidence-schema.md",{"path":123,"resourceType":124,"resourceTypeDesc":125,"sizeBytes":126,"sourceHtmlUrl":127},"scripts\u002Ftests\u002Ftest_verify_video_archive.py","SCRIPT","脚本",14160,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Fscripts\u002Ftests\u002Ftest_verify_video_archive.py",{"path":129,"resourceType":124,"resourceTypeDesc":125,"sizeBytes":130,"sourceHtmlUrl":131},"scripts\u002Fverify_video_archive.py",118991,"https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-video-archive\u002Fscripts\u002Fverify_video_archive.py",[133,140,147,154,161,168],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":139,"publishedAt":19},"01M7DEA67FREYEGQJYTWD9D0GB","writing-great-skills","Use when creating, modifying, evaluating a skill.","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-great-skills\u002FSKILL.md",[],[],{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":19},"01M7DEA3ZN566HSV6YJ8KPE824","writing-for-agents","Use when creating, editing, or reviewing a skill, AGENTS.md, CLAUDE.md, or another document agents read.","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":19},"01M7DEA0FRE0FEYV9XS0YTTBPJ","why","Use for questions about design rationale, the history of regressions or incidents, or the evidence behind thresholds and tradeoffs. Use `how` for runtime behavior. Do not use for plain commit or date lookups.","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwhy\u002FSKILL.md",[],[],{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":19},"01M7DE9VTG6ZEQH1M10J66T8MR","verify-transcript","Use when validating transcript CLI behavior, locating its verification features, or running the paid YouTube end-to-end check.","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-transcript\u002FSKILL.md",[],[],{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":167,"publishedAt":19},"01M7DE9RWQ6WNVG455HQFHVNFR","verify-skills","Use when verifying `just compile-skills`, `just remote-skills`, `just install-skills`, or `just skills-discover` in the skills repository, such as after changing their scripts or adding, renaming, or moving a skill.","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fverify-skills\u002FSKILL.md",[],[],{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":19},"01M7DE9QEE87P76327CRG2TH00","unslop","Use when communicating directly with the user or writing and editing documents.","https:\u002F\u002Fgithub.com\u002Fpascalandy\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Funslop\u002FSKILL.md",[],[]]