[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EHT7PMAZXBC3WRD74PAVG2":3,"skill-same:01M7EHT7PMAZXBC3WRD74PAVG2":68},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":15,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":47},"01M7EHT7PMAZXBC3WRD74PAVG2","anonymize-paper",null,"双盲匿名深扫 + camera-ready 干净还原——投稿前扫尽泄漏面（作者块\u002Fgrant 致谢\u002FPDF 元数据\u002F第一人称自引\u002F仓库与主页链接\u002FLaTeX 注释\u002F.bib 注释\u002F主目录路径\u002F补充材料），自带 stdlib 扫描脚本与可逆编辑清单，录用后一键还原作者信息；只建议不提交。触发：匿名化、双盲检查、有没有泄漏身份、去匿名、camera-ready 还原、anonymize、did I leak my identity。","deyi2026\u002Fllm-first-loop","deyi2026","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F267286635?v=4","01M7EHT3KGDPR2R5HCCRQA3HYM","01M7EHT3M140G7BHY5CBAJVRH4",1,"MIT","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fanonymize-paper\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fanonymize-paper\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: anonymize-paper\ndescription: 双盲匿名深扫 + camera-ready 干净还原——投稿前扫尽泄漏面（作者块\u002Fgrant 致谢\u002FPDF 元数据\u002F第一人称自引\u002F仓库与主页链接\u002FLaTeX 注释\u002F.bib 注释\u002F主目录路径\u002F补充材料），自带 stdlib 扫描脚本与可逆编辑清单，录用后一键还原作者信息；只建议不提交。触发：匿名化、双盲检查、有没有泄漏身份、去匿名、camera-ready 还原、anonymize、did I leak my identity。\nlicense: MIT\nmetadata:\n  vendored-from: ShaishavMaisuria\u002Fresearch-paper-lifecycle-skills@f9a0ed8\n  adapted-for: llm-first-loop（2026-10-10）\n---\n\n# Anonymize Paper\n\n> **LFL 本地化说明（2026-10-10）**：源自 ShaishavMaisuria\u002Fresearch-paper-lifecycle-skills@f9a0ed8f0f1770fb0321c5b022d55ed34c939927（MIT，LICENSE 随附）。scripts\u002Fscan_anonymization.py 纯 stdlib，本机已实测（与 preflight-check 的 anonymization 段结果一致，check-id 共享属实；独立运行支持 --venue\u002F--blind）。\n\nMake a LaTeX submission genuinely double-blind — then undo it cleanly after\nacceptance. Anonymization leaks are documented desk-reject grounds (CHI and\nNeurIPS state this explicitly, including leaks in supplementary material and\nlinked repos), and de-anonymization done by hand routinely leaves placeholder\nauthors or dead `anonymous.4open.science` links in the published PDF. This\nskill runs the deep sweep in both directions and records every change so the\nreversal is mechanical, not archaeological.\n\n## When to use\n\n- \"Anonymize my paper for NeurIPS \u002F CHI \u002F ICML \u002F KDD ...\" \u002F \"blind this\"\n- \"Did I leak my identity anywhere?\" \u002F \"double-blind check\" before submission\n- \"Rewrite my self-citations in third person\"\n- \"My code\u002Fdata links identify me — what do I do?\"\n- After acceptance: \"de-anonymize\", \"restore the authors\", \"prepare the\n  camera-ready author block\"\n- Called from `tailor-to-venue` (anonymization sweep step) or before\n  `preflight-check` (final gate).\n\n## Inputs\n\n1. The main `.tex` file (with `\\documentclass`); `\\input`\u002F`\\include` files\n   are followed automatically.\n2. The venue profile `venues\u002Fconferences\u002F\u003Cvenue>-\u003Cyear>.yml` (schema in\n   `venues\u002Fschema.yml`) — supplies the blind level (single\u002Fdouble\u002Ftriple) and\n   the `cfp_url`. No profile? Ask the user for the blind level or create a\n   profile with `parse-cfp`.\n3. Optional but recommended: the compiled PDF (metadata check), the\n   supplementary directory, and the author\u002Finstitution names to grep for.\n\n## Process — anonymize (submission)\n\n1. **Resolve the blind level, then re-verify it live — mandatory.** Read the\n   venue profile; fetch the `cfp_url` and confirm the blind level and the\n   venue's anonymization policy wording (what counts as a violation, whether\n   acknowledgments must be removed, whether anonymized artifact links are\n   allowed). Single-blind venues (e.g. SIGSPATIAL) need *no* anonymization —\n   tell the user and stop instead of mangling a fine paper.\n\n2. **Run the deep scan:**\n\n   ```\n   python3 scripts\u002Fscan_anonymization.py paper.tex \\\n       --venue venues\u002Fconferences\u002F\u003Cvenue>-\u003Cyear>.yml \\\n       --supplementary \u003Csupp-dir> --names \"Jane Doe,Example University\"\n   ```\n\n   Flags: `--blind double` (no profile), `--pdf paper.pdf` (explicit PDF),\n   `--no-pdf`, `--json`, `--strict` (warnings also fail), `--force` (scan at\n   single-blind venues anyway), `--no-inputs`. Exit codes: 0 clean, 1 leaks\n   found, 2 bad arguments. The scanner covers: author\u002Faffiliation\u002Femail\u002F\n   ORCID\u002F`\\thanks` blocks, acknowledgments, funding\u002Fgrant ids, identifying\n   links, bare emails, first-person self-citations, institutional\n   self-references, `pdfauthor`, LaTeX comments, `.bib` files, home-directory\n   paths, compiled-PDF metadata bytes, and the supplementary tree.\n\n3. **Fix findings with reversible edits.** Work through ERRORs first, then\n   WARNs (each is a judgment call — discuss, don't bulk-delete). For the fix\n   recipe per leak class — author block, acknowledgments, self-citations,\n   repo\u002Fdataset links, metadata, supplementary — follow\n   [references\u002Fleak-catalog.md](references\u002Fleak-catalog.md). Two rules:\n   - Prefer the `\\ifanon` toggle so the camera-ready flip is one line; fall\n     back to a `submission-anon` git branch when source will be uploaded\n     (arXiv, supplementary zips) — toggles leak the real names in source.\n   - Rewrite self-citations in third person (\"Doe et al. [12] showed\"), never\n     as \"Anonymous [12]\" — unless the cited work is itself unpublished.\n\n4. **Record the reversal manifest.** For every change write one line into\n   `anonymization-manifest.md` (kept OUT of the submission zip): what was\n   removed\u002Frewritten, file:line, and the exact original text (grant numbers,\n   acknowledgment paragraph, real repo URL). Format in\n   [references\u002Fcamera-ready-reversal.md](references\u002Fcamera-ready-reversal.md).\n\n5. **Sweep the supplementary material.** Re-run step 2 with `--supplementary`\n   after fixes. Ship code as a clean export (no `.git`), clear notebook\n   outputs\u002Fmetadata, remove LICENSE\u002FAUTHORS copyright names, and host\n   artifacts on an anonymized mirror (anonymous.4open.science) — never Drive\u002F\n   Dropbox\u002FGitHub links that expose the account.\n\n6. **Verify the compiled PDF.** Recompile, re-run the scan so the PDF\n   metadata check runs on the fresh PDF, and do the manual pass the scanner\n   cannot: figures with lab logos or terminal screenshots showing usernames,\n   dataset descriptions that name the institution, watermarks.\n\n7. **Re-run until clean, then gate.** Iterate scan → fix → scan to exit 0.\n   Then run the full `preflight-check` (it validates the documentclass\n   invocation and the rest of the desk-reject surface).\n\n## Process — de-anonymize (camera-ready)\n\n8. After acceptance, reverse using the manifest plus:\n\n   ```\n   python3 scripts\u002Fscan_anonymization.py paper.tex --mode camera-ready\n   ```\n\n   This flags the leftovers: placeholder\u002Fempty author blocks, lingering\n   `anonymous.4open.science` links, `[review,anonymous]` class options,\n   `neurips_\u003Cyear>` without `[final]`, `\\anontrue` toggles, \"omitted for\n   review\" wording, and missing acknowledgments. Restore the author block so\n   it matches the copyright form (ACM eRights \u002F IEEE eCF) EXACTLY — walk\n   [references\u002Fcamera-ready-reversal.md](references\u002Fcamera-ready-reversal.md)\n   — then hand off to `prepare-camera-ready` for the venue rail.\n\n## Output\n\n- A findings report (text or `--json`) with severity, check id, `file:line`.\n- The edited `.tex`\u002Fsupplementary files (with the user's approval, one leak\n  class at a time) plus `anonymization-manifest.md` for the reversal.\n- At camera-ready: a leftover report and the restored sources.\n\n## Relationship to preflight-check\n\n`preflight-check` runs the same source-level anonymization checks (shared\ncheck ids, `anonymization\u002F*`) as one gate among many; this skill is the deep\nvariant — comments, `.bib`, PDF bytes, supplementary trees, name grep — plus\nthe fix workflow and the camera-ready reversal. Quick gate → preflight; full\nsweep or de-anonymization → this skill.\n\n## Adapt to your discipline\n\nThe leak patterns are field-agnostic; the policies are not. Fork and swap the\nvenue profiles for your field's journals (many use single-blind — the scan\nthen auto-skips), and extend `--names` conventions for institutional review\nboards or clinical-trial registry ids that identify groups in your field.\n\n## Guardrails\n\n- Never claim the paper \"is anonymous\" — say \"no machine-detectable identity\n  leaks remain\"; writing style, self-datasets, and niche topics can still\n  identify authors, and reviewers actively search.\n- Never delete scholarly content to anonymize (e.g. dropping a self-citation\n  entirely is misconduct-adjacent); rewrite in third person instead. Citation\n  integrity questions route through `verify-citations`.\n- Re-verify the blind level and anonymization policy against the live\n  `cfp_url` before editing — a wrong blind level mangles a correct paper.\n- Never submit to any system on the user's behalf; stop at the report\u002Fedits.\n- Quote at most the flagged line in reports; never paste large paper\n  portions into outputs.\n","# Anonymize Paper\n\n> **LFL 本地化说明（2026-10-10）**：源自 ShaishavMaisuria\u002Fresearch-paper-lifecycle-skills@f9a0ed8f0f1770fb0321c5b022d55ed34c939927（MIT，LICENSE 随附）。scripts\u002Fscan_anonymization.py 纯 stdlib，本机已实测（与 preflight-check 的 anonymization 段结果一致，check-id 共享属实；独立运行支持 --venue\u002F--blind）。\n\nMake a LaTeX submission genuinely double-blind — then undo it cleanly after\nacceptance. Anonymization leaks are documented desk-reject grounds (CHI and\nNeurIPS state this explicitly, including leaks in supplementary material and\nlinked repos), and de-anonymization done by hand routinely leaves placeholder\nauthors or dead `anonymous.4open.science` links in the published PDF. This\nskill runs the deep sweep in both directions and records every change so the\nreversal is mechanical, not archaeological.\n\n## When to use\n\n- \"Anonymize my paper for NeurIPS \u002F CHI \u002F ICML \u002F KDD ...\" \u002F \"blind this\"\n- \"Did I leak my identity anywhere?\" \u002F \"double-blind check\" before submission\n- \"Rewrite my self-citations in third person\"\n- \"My code\u002Fdata links identify me — what do I do?\"\n- After acceptance: \"de-anonymize\", \"restore the authors\", \"prepare the\n  camera-ready author block\"\n- Called from `tailor-to-venue` (anonymization sweep step) or before\n  `preflight-check` (final gate).\n\n## Inputs\n\n1. The main `.tex` file (with `\\documentclass`); `\\input`\u002F`\\include` files\n   are followed automatically.\n2. The venue profile `venues\u002Fconferences\u002F\u003Cvenue>-\u003Cyear>.yml` (schema in\n   `venues\u002Fschema.yml`) — supplies the blind level (single\u002Fdouble\u002Ftriple) and\n   the `cfp_url`. No profile? Ask the user for the blind level or create a\n   profile with `parse-cfp`.\n3. Optional but recommended: the compiled PDF (metadata check), the\n   supplementary directory, and the author\u002Finstitution names to grep for.\n\n## Process — anonymize (submission)\n\n1. **Resolve the blind level, then re-verify it live — mandatory.** Read the\n   venue profile; fetch the `cfp_url` and confirm the blind level and the\n   venue's anonymization policy wording (what counts as a violation, whether\n   acknowledgments must be removed, whether anonymized artifact links are\n   allowed). Single-blind venues (e.g. SIGSPATIAL) need *no* anonymization —\n   tell the user and stop instead of mangling a fine paper.\n\n2. **Run the deep scan:**\n\n   ```\n   python3 scripts\u002Fscan_anonymization.py paper.tex \\\n       --venue venues\u002Fconferences\u002F\u003Cvenue>-\u003Cyear>.yml \\\n       --supplementary \u003Csupp-dir> --names \"Jane Doe,Example University\"\n   ```\n\n   Flags: `--blind double` (no profile), `--pdf paper.pdf` (explicit PDF),\n   `--no-pdf`, `--json`, `--strict` (warnings also fail), `--force` (scan at\n   single-blind venues anyway), `--no-inputs`. Exit codes: 0 clean, 1 leaks\n   found, 2 bad arguments. The scanner covers: author\u002Faffiliation\u002Femail\u002F\n   ORCID\u002F`\\thanks` blocks, acknowledgments, funding\u002Fgrant ids, identifying\n   links, bare emails, first-person self-citations, institutional\n   self-references, `pdfauthor`, LaTeX comments, `.bib` files, home-directory\n   paths, compiled-PDF metadata bytes, and the supplementary tree.\n\n3. **Fix findings with reversible edits.** Work through ERRORs first, then\n   WARNs (each is a judgment call — discuss, don't bulk-delete). For the fix\n   recipe per leak class — author block, acknowledgments, self-citations,\n   repo\u002Fdataset links, metadata, supplementary — follow\n   [references\u002Fleak-catalog.md](references\u002Fleak-catalog.md). Two rules:\n   - Prefer the `\\ifanon` toggle so the camera-ready flip is one line; fall\n     back to a `submission-anon` git branch when source will be uploaded\n     (arXiv, supplementary zips) — toggles leak the real names in source.\n   - Rewrite self-citations in third person (\"Doe et al. [12] showed\"), never\n     as \"Anonymous [12]\" — unless the cited work is itself unpublished.\n\n4. **Record the reversal manifest.** For every change write one line into\n   `anonymization-manifest.md` (kept OUT of the submission zip): what was\n   removed\u002Frewritten, file:line, and the exact original text (grant numbers,\n   acknowledgment paragraph, real repo URL). Format in\n   [references\u002Fcamera-ready-reversal.md](references\u002Fcamera-ready-reversal.md).\n\n5. **Sweep the supplementary material.** Re-run step 2 with `--supplementary`\n   after fixes. Ship code as a clean export (no `.git`), clear notebook\n   outputs\u002Fmetadata, remove LICENSE\u002FAUTHORS copyright names, and host\n   artifacts on an anonymized mirror (anonymous.4open.science) — never Drive\u002F\n   Dropbox\u002FGitHub links that expose the account.\n\n6. **Verify the compiled PDF.** Recompile, re-run the scan so the PDF\n   metadata check runs on the fresh PDF, and do the manual pass the scanner\n   cannot: figures with lab logos or terminal screenshots showing usernames,\n   dataset descriptions that name the institution, watermarks.\n\n7. **Re-run until clean, then gate.** Iterate scan → fix → scan to exit 0.\n   Then run the full `preflight-check` (it validates the documentclass\n   invocation and the rest of the desk-reject surface).\n\n## Process — de-anonymize (camera-ready)\n\n8. After acceptance, reverse using the manifest plus:\n\n   ```\n   python3 scripts\u002Fscan_anonymization.py paper.tex --mode camera-ready\n   ```\n\n   This flags the leftovers: placeholder\u002Fempty author blocks, lingering\n   `anonymous.4open.science` links, `[review,anonymous]` class options,\n   `neurips_\u003Cyear>` without `[final]`, `\\anontrue` toggles, \"omitted for\n   review\" wording, and missing acknowledgments. Restore the author block so\n   it matches the copyright form (ACM eRights \u002F IEEE eCF) EXACTLY — walk\n   [references\u002Fcamera-ready-reversal.md](references\u002Fcamera-ready-reversal.md)\n   — then hand off to `prepare-camera-ready` for the venue rail.\n\n## Output\n\n- A findings report (text or `--json`) with severity, check id, `file:line`.\n- The edited `.tex`\u002Fsupplementary files (with the user's approval, one leak\n  class at a time) plus `anonymization-manifest.md` for the reversal.\n- At camera-ready: a leftover report and the restored sources.\n\n## Relationship to preflight-check\n\n`preflight-check` runs the same source-level anonymization checks (shared\ncheck ids, `anonymization\u002F*`) as one gate among many; this skill is the deep\nvariant — comments, `.bib`, PDF bytes, supplementary trees, name grep — plus\nthe fix workflow and the camera-ready reversal. Quick gate → preflight; full\nsweep or de-anonymization → this skill.\n\n## Adapt to your discipline\n\nThe leak patterns are field-agnostic; the policies are not. Fork and swap the\nvenue profiles for your field's journals (many use single-blind — the scan\nthen auto-skips), and extend `--names` conventions for institutional review\nboards or clinical-trial registry ids that identify groups in your field.\n\n## Guardrails\n\n- Never claim the paper \"is anonymous\" — say \"no machine-detectable identity\n  leaks remain\"; writing style, self-datasets, and niche topics can still\n  identify authors, and reviewers actively search.\n- Never delete scholarly content to anonymize (e.g. dropping a self-citation\n  entirely is misconduct-adjacent); rewrite in third person instead. Citation\n  integrity questions route through `verify-citations`.\n- Re-verify the blind level and anonymization policy against the live\n  `cfp_url` before editing — a wrong blind level mangles a correct paper.\n- Never submit to any system on the user's behalf; stop at the report\u002Fedits.\n- Quote at most the flagged line in reports; never paste large paper\n  portions into outputs.\n",{"name":5,"license":14,"metadata":24,"description":7},{"adapted-for":25,"vendored-from":26},"llm-first-loop（2026-10-10）","ShaishavMaisuria\u002Fresearch-paper-lifecycle-skills@f9a0ed8","main","57f220f344f02e6abf6eb55434077d104c209804","skills\u002Fanonymize-paper",0,"https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop","https:\u002F\u002Fgithub.com\u002Fdeyi2026","2026-11-15 10:23:32",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add deyi2026\u002Fllm-first-loop --skill anonymize-paper",{"tool":40,"command":41},"bunx","bunx skills add deyi2026\u002Fllm-first-loop --skill anonymize-paper",{"tool":43,"command":44},"pnpm","pnpm dlx skills add deyi2026\u002Fllm-first-loop --skill anonymize-paper","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fanonymize-paper\u002FSKILL.md",true,[48,52,58,62],{"path":49,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":15},"LICENSE","许可证",10278,{"path":53,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":57},"references\u002Fcamera-ready-reversal.md","REFERENCE","参考文档",7148,"https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fanonymize-paper\u002Freferences\u002Fcamera-ready-reversal.md",{"path":59,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":60,"sourceHtmlUrl":61},"references\u002Fleak-catalog.md",9147,"https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fanonymize-paper\u002Freferences\u002Fleak-catalog.md",{"path":63,"resourceType":64,"resourceTypeDesc":65,"sizeBytes":66,"sourceHtmlUrl":67},"scripts\u002Fscan_anonymization.py","SCRIPT","脚本",41928,"https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fanonymize-paper\u002Fscripts\u002Fscan_anonymization.py",[69,77,84,91,98,105],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7EJ01AM5PZXYNM9GBF6CW12","security-audit","Audit project for security issues.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fvendor_gemini_demo\u002Fskills\u002Fsecurity-audit\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7EJ00CD2JXGVB0JETXGEX43","metrics","Collect project metrics.","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fvendor_codex_demo\u002Fskills\u002Fmetrics\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7EHZZD7DVVTSECGF5HHHQZ5","lint","Lint the repo using project config.","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fvendor_claude_demo\u002Fskills\u002Flint\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":19},"01M7EHZYGGYKDZDEHVF4YSD22X","report","值班报告生成","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fstd_plugin_demo\u002Fskills\u002Freport\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M7EHZVX0V6ER3X3017K47RK4","deploy","标准部署流程（多行 YAML frontmatter 演示）","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fstd_plugin_demo\u002Fskills\u002Fdeploy\u002FSKILL.md",[],[],{"skillNo":106,"name":107,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":73,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":19},"01M7EHZRENGPKXXFBA3YJ1QF1S","web-self-restart-triage","https:\u002F\u002Fgithub.com\u002Fdeyi2026\u002Fllm-first-loop\u002Fblob\u002Fmain\u002Fskills\u002Fweb-self-restart-triage\u002FSKILL.md",[],[]]