[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71FNJ6S5MATHHD81PMD42MM":3,"skill-same:01M71FNJ6S5MATHHD81PMD42MM":70},{"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":37,"linkOnly":38,"markdownContent":39,"markdownBody":40,"frontMatter":41,"refName":42,"commitSha":43,"skillDirectory":44,"forksCount":45,"repositoryHtmlUrl":46,"repositoryHomepageUrl":6,"ownerHtmlUrl":47,"ownerBlogUrl":48,"ownerDescription":6,"updatedAt":49,"install":50,"resources":63},"01M71FNJ6S5MATHHD81PMD42MM","upload-parity-experiments",null,"Create or reuse Hugging Face dataset PRs for `harborframework\u002Fparity-experiments` and upload Harbor parity\u002Foracle result folders efficiently with sparse checkout, raw git pushes, and Git LFS.","pku-liang\u002Fhwe-bench","pku-liang","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F81861363?v=4","01M71FMJQ2VZ4YEERAS5FJ438T","PKU Yun (Eric) Liang Research Group","01M71FMJR6AXWMX6K35KS4ZS6G",73,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fupload-parity-experiments\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36],"upload parity experiments","experiments","parity","upload","face","hugging face","git","result","create","with","harbor","dataset","oracle","reuse","prs","lfs","for","checkout","2026-11-10 10:12:30",false,"---\nname: upload-parity-experiments\ndescription: Create or reuse Hugging Face dataset PRs for `harborframework\u002Fparity-experiments` and upload Harbor parity\u002Foracle result folders efficiently with sparse checkout, raw git pushes, and Git LFS.\n---\n\n# Upload Parity Experiments\n\nUse this skill to publish Harbor parity experiment outputs to the shared Hugging Face dataset and capture the resulting discussion URL for the adapter's `parity_pr` field.\n\n## Why This Skill Exists\n\n- `hf upload-large-folder` can be slow or unreliable for large parity bundles because it pushes through the Hub API commit loop.\n- A normal git clone of `harborframework\u002Fparity-experiments` is too expensive because the dataset is very large.\n- Hugging Face dataset PR refs are different from GitHub PR refs and are easy to misuse.\n- Files larger than 10 MiB must be Git LFS-tracked before pushing.\n\nThis skill avoids the full clone by fetching only the target PR ref with `--depth 1 --filter=blob:none` and checking out only the paths needed for the current adapter.\n\n## Prereqs\n\n- Ensure Hugging Face authentication is available with discussion-write permission. Either a classic `write` token or a fine-grained token with global `discussion.write` enabled at \u003Chttps:\u002F\u002Fhuggingface.co\u002Fsettings\u002Ftokens>. A read-only or narrowly-scoped token will cause `create_pr.py` to fail with HTTP 403.\n- Keep the target dataset fixed to `harborframework\u002Fparity-experiments` unless the user explicitly asks for another repo.\n- Accept any local upload source that already contains the final files the user wants to publish.\n\n## Preferred Workflow\n\n1. Create or reuse a dataset PR.\n2. Prepare a sparse local worktree for that PR ref.\n3. Copy the local parity results into the sparse checkout.\n4. Ensure every file larger than 10 MiB is Git LFS-tracked before committing.\n5. Push directly to the PR ref with raw `git push`.\n6. Share the discussion URL and record it as the adapter's `parity_pr`.\n\nFor large parity bundles, prefer raw git over `hf upload-large-folder`. The raw git path is materially faster and more reliable because it avoids the API-side commit loop and does not require cloning the entire parity dataset.\n\n## 1. Create Or Reuse A Dataset PR\n\nIf the user already has a parity PR number, reuse it.\n\nOtherwise, create one with the bundled helper:\n\n```bash\nuv run python scripts\u002Fcreate_pr.py create-pr \\\n  --title \"Add parity experiments for \u003Cadapter_name>\" \\\n  --description-file \u002Fpath\u002Fto\u002Fpr-description.md\n```\n\nThe script prints JSON including:\n\n- `pr_number`\n- `discussion_url`\n- `repo_id`\n\n## 2. Prepare A Sparse PR Checkout\n\n```bash\nmkdir -p \u002Ftmp\u002Fparity-experiments-pr\u003Cnumber>\ncd \u002Ftmp\u002Fparity-experiments-pr\u003Cnumber>\n\ngit init\ngit remote add origin git@hf.co:datasets\u002Fharborframework\u002Fparity-experiments\ngit config core.sparseCheckout true\ngit sparse-checkout init --cone\ngit sparse-checkout set adapters\u002F\u003Cadapter_name>\n\ngit fetch --depth 1 --filter=blob:none origin refs\u002Fpr\u002F\u003Cnumber>:pr\u002F\u003Cnumber>\ngit checkout pr\u002F\u003Cnumber>\n```\n\nThis fetches only the PR ref and the requested paths instead of cloning the full dataset repo.\n\n## 3. Copy The Local Results\n\nIf the local folder already contains the final repo-root layout, copy it as-is.\n\nIf the local folder only contains the adapter subtree, copy it into `adapters\u002F\u003Cadapter_name>\u002F`:\n\n```bash\nrsync -a --delete \\\n  --exclude '.git' \\\n  --exclude '.cache' \\\n  --exclude '.DS_Store' \\\n  \u002Fpath\u002Fto\u002Flocal-folder\u002F \\\n  adapters\u002F\u003Cadapter_name>\u002F\n```\n\n## 4. Ensure Large Files Use Git LFS\n\nThe repo-root `.gitattributes` already LFS-tracks common binary, model, archive, and media extensions (`*.bin`, `*.parquet`, `*.safetensors`, images, audio, video, `*.log`, `*.txt`, etc.). Most parity outputs are covered automatically and need no manual action. Scan the sparse checkout for files larger than 10 MiB to catch anything that slipped through:\n\n```bash\npython - \u003C\u003C'PY'\nfrom pathlib import Path\n\nfor path in sorted(Path(\".\").rglob(\"*\")):\n    if path.is_file() and \".git\" not in path.parts and path.stat().st_size > 10 * 1024 * 1024:\n        print(path)\nPY\n```\n\nIf any file is flagged and is not already covered by root, write the LFS rule to `adapters\u002F\u003Cadapter_name>\u002F.gitattributes` — never to the repo-root `.gitattributes`, which is a shared merge-conflict hotspot. Run `git lfs track` inside the adapter directory so the rule is written with a relative pattern:\n\n```bash\n(cd adapters\u002F\u003Cadapter_name> && git lfs track \"\u003Cpattern>\")\ngit add adapters\u002F\u003Cadapter_name>\u002F.gitattributes\n```\n\nGit LFS honors nested `.gitattributes` files, so rules added this way apply only to that adapter and never collide with other in-flight parity PRs.\n\n## 5. Commit And Push\n\n```bash\nfind . -name .DS_Store -delete\ngit add adapters\u002F\u003Cadapter_name>\ngit commit -m \"Add parity experiment artifacts for \u003Cadapter_name>\"\nGIT_SSH_COMMAND='ssh -o ServerAliveInterval=30 -o ServerAliveCountMax=10' \\\n  git push origin pr\u002F\u003Cnumber>:refs\u002Fpr\u002F\u003Cnumber>\n```\n\n## 6. Record The Discussion URL\n\nUse this discussion URL in `parity_experiment.json`:\n\n```text\nhttps:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fharborframework\u002Fparity-experiments\u002Fdiscussions\u002F\u003Cnumber>\n```\n\n## Monitoring And Retry\n\nWhen raw `git push` is running, check the live output first. Large pushes often spend a long time in:\n\n- `git-lfs pre-push`\n- single-object LFS uploads\n- packfile compression\n\nUseful checks:\n\n```bash\ngit ls-remote git@hf.co:datasets\u002Fharborframework\u002Fparity-experiments refs\u002Fpr\u002F\u003Cnumber>\n```\n\n```bash\nps -p \u003Cpid> -o pid=,etime=,command=\n```\n\nIf the push exits early:\n\n- rerun the exact same `git push` command first\n- if Hugging Face rejects a file larger than 10 MiB, add the needed `git lfs track` rules, re-add the file, and retry\n- if the connection broke mid-push, rerun the same push command before changing anything else\n\n## Fallback\n\nUse `hf upload-large-folder` only as a fallback when raw git is unavailable or the user explicitly asks for it.\n\nYou can still create the PR with the bundled helper and upload to the PR revision:\n\n```bash\nhf upload-large-folder harborframework\u002Fparity-experiments \\\n  \u003Clocal-folder> \\\n  --repo-type dataset \\\n  --revision refs\u002Fpr\u002F\u003Cpr-number> \\\n  --exclude \".DS_Store\" \\\n  --exclude \"**\u002F.DS_Store\" \\\n  --num-workers 8\n```\n\n## Guardrails\n\n- Do not upload the Harbor repo itself by accident. Upload only the intended local results folder.\n- Do not fall back to a full clone of `harborframework\u002Fparity-experiments`.\n- Do not modify the repo-root `.gitattributes`. Put any adapter-specific LFS rules in `adapters\u002F\u003Cadapter_name>\u002F.gitattributes` instead.\n- Always remove or exclude `.DS_Store`.\n- Before pushing, make sure every file larger than 10 MiB is LFS-tracked.\n- If the user already has a parity PR number, reuse it instead of creating another one.\n","# Upload Parity Experiments\n\nUse this skill to publish Harbor parity experiment outputs to the shared Hugging Face dataset and capture the resulting discussion URL for the adapter's `parity_pr` field.\n\n## Why This Skill Exists\n\n- `hf upload-large-folder` can be slow or unreliable for large parity bundles because it pushes through the Hub API commit loop.\n- A normal git clone of `harborframework\u002Fparity-experiments` is too expensive because the dataset is very large.\n- Hugging Face dataset PR refs are different from GitHub PR refs and are easy to misuse.\n- Files larger than 10 MiB must be Git LFS-tracked before pushing.\n\nThis skill avoids the full clone by fetching only the target PR ref with `--depth 1 --filter=blob:none` and checking out only the paths needed for the current adapter.\n\n## Prereqs\n\n- Ensure Hugging Face authentication is available with discussion-write permission. Either a classic `write` token or a fine-grained token with global `discussion.write` enabled at \u003Chttps:\u002F\u002Fhuggingface.co\u002Fsettings\u002Ftokens>. A read-only or narrowly-scoped token will cause `create_pr.py` to fail with HTTP 403.\n- Keep the target dataset fixed to `harborframework\u002Fparity-experiments` unless the user explicitly asks for another repo.\n- Accept any local upload source that already contains the final files the user wants to publish.\n\n## Preferred Workflow\n\n1. Create or reuse a dataset PR.\n2. Prepare a sparse local worktree for that PR ref.\n3. Copy the local parity results into the sparse checkout.\n4. Ensure every file larger than 10 MiB is Git LFS-tracked before committing.\n5. Push directly to the PR ref with raw `git push`.\n6. Share the discussion URL and record it as the adapter's `parity_pr`.\n\nFor large parity bundles, prefer raw git over `hf upload-large-folder`. The raw git path is materially faster and more reliable because it avoids the API-side commit loop and does not require cloning the entire parity dataset.\n\n## 1. Create Or Reuse A Dataset PR\n\nIf the user already has a parity PR number, reuse it.\n\nOtherwise, create one with the bundled helper:\n\n```bash\nuv run python scripts\u002Fcreate_pr.py create-pr \\\n  --title \"Add parity experiments for \u003Cadapter_name>\" \\\n  --description-file \u002Fpath\u002Fto\u002Fpr-description.md\n```\n\nThe script prints JSON including:\n\n- `pr_number`\n- `discussion_url`\n- `repo_id`\n\n## 2. Prepare A Sparse PR Checkout\n\n```bash\nmkdir -p \u002Ftmp\u002Fparity-experiments-pr\u003Cnumber>\ncd \u002Ftmp\u002Fparity-experiments-pr\u003Cnumber>\n\ngit init\ngit remote add origin git@hf.co:datasets\u002Fharborframework\u002Fparity-experiments\ngit config core.sparseCheckout true\ngit sparse-checkout init --cone\ngit sparse-checkout set adapters\u002F\u003Cadapter_name>\n\ngit fetch --depth 1 --filter=blob:none origin refs\u002Fpr\u002F\u003Cnumber>:pr\u002F\u003Cnumber>\ngit checkout pr\u002F\u003Cnumber>\n```\n\nThis fetches only the PR ref and the requested paths instead of cloning the full dataset repo.\n\n## 3. Copy The Local Results\n\nIf the local folder already contains the final repo-root layout, copy it as-is.\n\nIf the local folder only contains the adapter subtree, copy it into `adapters\u002F\u003Cadapter_name>\u002F`:\n\n```bash\nrsync -a --delete \\\n  --exclude '.git' \\\n  --exclude '.cache' \\\n  --exclude '.DS_Store' \\\n  \u002Fpath\u002Fto\u002Flocal-folder\u002F \\\n  adapters\u002F\u003Cadapter_name>\u002F\n```\n\n## 4. Ensure Large Files Use Git LFS\n\nThe repo-root `.gitattributes` already LFS-tracks common binary, model, archive, and media extensions (`*.bin`, `*.parquet`, `*.safetensors`, images, audio, video, `*.log`, `*.txt`, etc.). Most parity outputs are covered automatically and need no manual action. Scan the sparse checkout for files larger than 10 MiB to catch anything that slipped through:\n\n```bash\npython - \u003C\u003C'PY'\nfrom pathlib import Path\n\nfor path in sorted(Path(\".\").rglob(\"*\")):\n    if path.is_file() and \".git\" not in path.parts and path.stat().st_size > 10 * 1024 * 1024:\n        print(path)\nPY\n```\n\nIf any file is flagged and is not already covered by root, write the LFS rule to `adapters\u002F\u003Cadapter_name>\u002F.gitattributes` — never to the repo-root `.gitattributes`, which is a shared merge-conflict hotspot. Run `git lfs track` inside the adapter directory so the rule is written with a relative pattern:\n\n```bash\n(cd adapters\u002F\u003Cadapter_name> && git lfs track \"\u003Cpattern>\")\ngit add adapters\u002F\u003Cadapter_name>\u002F.gitattributes\n```\n\nGit LFS honors nested `.gitattributes` files, so rules added this way apply only to that adapter and never collide with other in-flight parity PRs.\n\n## 5. Commit And Push\n\n```bash\nfind . -name .DS_Store -delete\ngit add adapters\u002F\u003Cadapter_name>\ngit commit -m \"Add parity experiment artifacts for \u003Cadapter_name>\"\nGIT_SSH_COMMAND='ssh -o ServerAliveInterval=30 -o ServerAliveCountMax=10' \\\n  git push origin pr\u002F\u003Cnumber>:refs\u002Fpr\u002F\u003Cnumber>\n```\n\n## 6. Record The Discussion URL\n\nUse this discussion URL in `parity_experiment.json`:\n\n```text\nhttps:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fharborframework\u002Fparity-experiments\u002Fdiscussions\u002F\u003Cnumber>\n```\n\n## Monitoring And Retry\n\nWhen raw `git push` is running, check the live output first. Large pushes often spend a long time in:\n\n- `git-lfs pre-push`\n- single-object LFS uploads\n- packfile compression\n\nUseful checks:\n\n```bash\ngit ls-remote git@hf.co:datasets\u002Fharborframework\u002Fparity-experiments refs\u002Fpr\u002F\u003Cnumber>\n```\n\n```bash\nps -p \u003Cpid> -o pid=,etime=,command=\n```\n\nIf the push exits early:\n\n- rerun the exact same `git push` command first\n- if Hugging Face rejects a file larger than 10 MiB, add the needed `git lfs track` rules, re-add the file, and retry\n- if the connection broke mid-push, rerun the same push command before changing anything else\n\n## Fallback\n\nUse `hf upload-large-folder` only as a fallback when raw git is unavailable or the user explicitly asks for it.\n\nYou can still create the PR with the bundled helper and upload to the PR revision:\n\n```bash\nhf upload-large-folder harborframework\u002Fparity-experiments \\\n  \u003Clocal-folder> \\\n  --repo-type dataset \\\n  --revision refs\u002Fpr\u002F\u003Cpr-number> \\\n  --exclude \".DS_Store\" \\\n  --exclude \"**\u002F.DS_Store\" \\\n  --num-workers 8\n```\n\n## Guardrails\n\n- Do not upload the Harbor repo itself by accident. Upload only the intended local results folder.\n- Do not fall back to a full clone of `harborframework\u002Fparity-experiments`.\n- Do not modify the repo-root `.gitattributes`. Put any adapter-specific LFS rules in `adapters\u002F\u003Cadapter_name>\u002F.gitattributes` instead.\n- Always remove or exclude `.DS_Store`.\n- Before pushing, make sure every file larger than 10 MiB is LFS-tracked.\n- If the user already has a parity PR number, reuse it instead of creating another one.\n",{"name":5,"description":7},"main","10c78a87e1f92695d78d15b1464a6107dcac8837","deps\u002Fharbor\u002Fskills\u002Fupload-parity-experiments",9,"https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench","https:\u002F\u002Fgithub.com\u002Fpku-liang","https:\u002F\u002Fericlyun.github.io\u002F","2026-08-02 01:13:52",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add pku-liang\u002Fhwe-bench --skill upload-parity-experiments",{"tool":56,"command":57},"bunx","bunx skills add pku-liang\u002Fhwe-bench --skill upload-parity-experiments",{"tool":59,"command":60},"pnpm","pnpm dlx skills add pku-liang\u002Fhwe-bench --skill upload-parity-experiments","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fupload-parity-experiments\u002FSKILL.md",true,[64],{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fcreate_pr.py","SCRIPT","脚本",2171,"https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fupload-parity-experiments\u002Fscripts\u002Fcreate_pr.py",[71,91,105,127,147,169],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":37},"01M71FNG0TQD738NZGZN1W31AF","rewardkit","Write Harbor task verifiers using Reward Kit. Use when creating or editing a task's tests\u002F directory, adding grading criteria, setting up LLM\u002Fagent judges, or designing verifiers that produce a reward score.","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Frewardkit\u002FSKILL.md",[],[73,78,79,80,81,82,29,83,84,85,86,87,88,89,90],"grading","tests","write","directory","Verifiers","task","when","use","produce","kit","up","editing","setting",{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":37},"01M71FNCKK74MFE7MYSPVN0YAT","publish","Publish a Harbor task or dataset to the registry. Use when the user wants to upload, publish, or share tasks or datasets\u002Fbenchmarks on the Harbor registry.","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fpublish\u002FSKILL.md",[],[93,98,30,84,85,99,100,22,101,29,102,83,103,104],"tasks","registry","datasets","share","path","user","benchmarks",{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":37},"01M71FN9MJMRMA5E1AZCRFKN6M","harbor-exec","Use when working with Harbor's `harbor exec` CLI workflow: compiling files, directories, or globs into Harbor tasks; running map jobs; configuring artifacts and existence-only verification; using map-reduce; writing or reviewing ExecConfig YAML\u002FJSON\u002FTOML; or debugging command behavior, config validation, and job outputs.","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fharbor-exec\u002FSKILL.md",[],[112,29,113,85,114,115,28,98,116,117,118,119,84,120,121,122,123,124,125,126],"harbor exec","exec","writing","verification","artifacts","debugging","workflow","reduce","map","only","outputs","toml","directories","running","files",{"skillNo":128,"name":129,"nameZh":6,"description":130,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":131,"categories":132,"keywords":133,"publishedAt":37},"01M71FN6EG0E0TG5HVC5ZGV6PW","create-task","Create a new Harbor task for evaluating agents. Use when the user wants to scaffold, build, or design a new task, benchmark problem, or eval. Guides through instruction writing, environment setup, verifier design (pytest vs Reward Kit vs custom), and solution scripting.","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fcreate-task\u002FSKILL.md",[],[134,83,27,135,136,137,138,139,85,140,141,35,142,143,144,145,84,114,146,87],"create task","eval","new task","problem","design","environment","custom","verifier","pytest","org","benchmark","guides","new",{"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":158,"publishedAt":37},"01M71FN46QH5B25X9SXCJ5PNWD","create-adapter","Scaffold a new Harbor benchmark adapter by running `harbor adapter init` and then guide implementation using the Adapters Agent Guide as the authoritative spec.","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fskills\u002Fcreate-adapter\u002FSKILL.md",[153],{"categoryNo":154,"categoryName":155,"taxonomyType":156,"taxonomyTypeDesc":157},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[159,160,27,161,162,163,164,165,166,144,29,146,167,168,125],"create adapter","adapter","spec","scaffold","guide","agent guide","Adapters","implementation","init","as",{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":173,"categories":174,"keywords":175,"publishedAt":37},"01M71FN35PFRFKD7ZKMH9WHZDV","generate-greeting","Generate a greeting message and write it to a file.","https:\u002F\u002Fgithub.com\u002Fpku-liang\u002Fhwe-bench\u002Fblob\u002Fmain\u002Fdeps\u002Fharbor\u002Fexamples\u002Ftasks\u002Fhello-skills\u002Fenvironment\u002Fskills\u002Fgenerate-greeting\u002FSKILL.md",[],[176,177,178,80,179,180,181],"greeting","generate","generate greeting","it","file","message"]