[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6AC5QAF7ZST45DS0PEKW9JK":3,"skill-same:01M6AC5QAF7ZST45DS0PEKW9JK":56},{"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":22,"linkOnly":23,"markdownContent":24,"markdownBody":25,"frontMatter":26,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":32,"ownerHtmlUrl":33,"ownerBlogUrl":34,"ownerDescription":6,"updatedAt":35,"install":36,"resources":49},"01M6AC5QAF7ZST45DS0PEKW9JK","implementation-kickoff",null,"Start or resume a requested Guardrails implementation or PR takeover in the selected linked worktree with bounded scope and verification.","openai\u002Fopenai-guardrails-js","openai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F14957082?v=4","01M6AC29YGCR4S2NTJFV0G8B81","OpenAI","01M6AC5JPN01DTSJMFDT21GDMJ",114,"MIT","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fimplementation-kickoff\u002FSKILL.md",[],[19,20,21],"implementation kickoff","pr","verification","2026-11-10 10:12:30",false,"---\nname: implementation-kickoff\ndescription: Start or resume a requested Guardrails implementation or PR takeover in the selected linked worktree with bounded scope and verification.\n---\n\n# Implementation Kickoff\n\nStart from the user's actual implementation request. An investigation or a plan\nalone is not permission to implement, push, or publish. Preserve authorization\nalready given; this skill adds none. Apply [AGENTS.md](..\u002F..\u002F..\u002FAGENTS.md) and\n[implementation-strategy](..\u002Fimplementation-strategy\u002FSKILL.md) where applicable.\n\n## Establish the checkout and scope\n\nRecord the original outcome, acceptance criteria, affected paths, non-goals,\nand intended base. Reuse the current task's selected linked worktree. If running\nin a primary checkout, follow AGENTS.md and ask to start or hand off to a worktree.\nDo not automatically create another worktree or change unrelated checkouts.\n\nRefresh remote references and record the full intended starting SHA. For a new\nindependent task use the refreshed default branch unless the user chose another\nbase. For an explicitly requested stack, use the exact previous slice's tip.\nVerify the checkout equals that SHA before editing a new task. For a resumed\ntask, preserve its existing changes and record HEAD plus the original comparison\nbase instead of resetting it to main. Stop on an unexplained base mismatch.\n\nInspect tracked and untracked state before edits. Keep task deliverables distinct\nfrom existing user work and operational plans\u002Freview notes. Use `codex\u002F` branch\nnames by default, preserving an explicitly selected branch or user preference.\n\n## Take over an existing PR\n\nRead current metadata, author commits, discussions, and the complete PR diff.\nConfirm the PR is still an appropriate source. Preserve its underlying outcome,\nbut evaluate the proposal against the current architecture and toolchain.\nImport only the accepted scope; a takeover does not require copying the entire\npatch or preserving intermediate commits.\n\nRecord the source PR and verified author\u002Fcoauthor identities. Credit adapted\nwork with valid `Co-authored-by` trailers, without inventing identities. If the\nidentity needed for attribution cannot be verified, resolve it before committing.\nFor a split takeover, describe each slice as part of the replacement effort;\ndo not claim the first slice completes the whole source PR.\n\n## Implement and verify\n\nKeep changes within the agreed scope. Preserve user work and avoid automatic\nstashing, rebasing, or rewriting unrelated branches. Check integration with an\nadvanced target before submission; if an update is needed, handle only this\ntask's branch and rerun checks\u002Freview affected by any content change. Do not\nforce a detached checkout, one-commit topology, or history rewrite to manufacture\na clean handoff.\n\nUse [verification](..\u002Fcode-change-verification\u002FSKILL.md) and\n[final review](..\u002Fimplementation-final-review\u002FSKILL.md) before pushing or creating\nor updating a PR. Follow the [release guide](..\u002F..\u002F..\u002F.changeset\u002FREADME.md):\nordinary user-visible SDK changes need a changeset; internal workflow changes\nneed none. Leave version and changelog generation to the release PR.\n\n## Handoff\n\nUse [pr-draft-summary](..\u002Fpr-draft-summary\u002FSKILL.md) for the final description\nand authorized submission\u002Ffollow-up.\n\nInspect status, full diff statistics, and every task commit before staging or\nreporting. Stage only task-owned deliverables; keep local planning and review\nartifacts out of the PR. Preserve hooks and inspect the committed result.\n\nReport branch, base, commit, requested outcome, verification, review status, and\nremaining blockers. When authorized to submit, follow AGENTS.md for current-head\nCI monitoring, scoped fixes, root-channel review requests, and feedback replies.\nFor a requested stack, create each PR against its predecessor and verify that\nrelationship. Do not assume all workflows run on feature-branch PR targets:\ninspect triggers and use an authorized manual CI run on each head when needed.\nNever merge or close the source PR without authorization.\n","# Implementation Kickoff\n\nStart from the user's actual implementation request. An investigation or a plan\nalone is not permission to implement, push, or publish. Preserve authorization\nalready given; this skill adds none. Apply [AGENTS.md](..\u002F..\u002F..\u002FAGENTS.md) and\n[implementation-strategy](..\u002Fimplementation-strategy\u002FSKILL.md) where applicable.\n\n## Establish the checkout and scope\n\nRecord the original outcome, acceptance criteria, affected paths, non-goals,\nand intended base. Reuse the current task's selected linked worktree. If running\nin a primary checkout, follow AGENTS.md and ask to start or hand off to a worktree.\nDo not automatically create another worktree or change unrelated checkouts.\n\nRefresh remote references and record the full intended starting SHA. For a new\nindependent task use the refreshed default branch unless the user chose another\nbase. For an explicitly requested stack, use the exact previous slice's tip.\nVerify the checkout equals that SHA before editing a new task. For a resumed\ntask, preserve its existing changes and record HEAD plus the original comparison\nbase instead of resetting it to main. Stop on an unexplained base mismatch.\n\nInspect tracked and untracked state before edits. Keep task deliverables distinct\nfrom existing user work and operational plans\u002Freview notes. Use `codex\u002F` branch\nnames by default, preserving an explicitly selected branch or user preference.\n\n## Take over an existing PR\n\nRead current metadata, author commits, discussions, and the complete PR diff.\nConfirm the PR is still an appropriate source. Preserve its underlying outcome,\nbut evaluate the proposal against the current architecture and toolchain.\nImport only the accepted scope; a takeover does not require copying the entire\npatch or preserving intermediate commits.\n\nRecord the source PR and verified author\u002Fcoauthor identities. Credit adapted\nwork with valid `Co-authored-by` trailers, without inventing identities. If the\nidentity needed for attribution cannot be verified, resolve it before committing.\nFor a split takeover, describe each slice as part of the replacement effort;\ndo not claim the first slice completes the whole source PR.\n\n## Implement and verify\n\nKeep changes within the agreed scope. Preserve user work and avoid automatic\nstashing, rebasing, or rewriting unrelated branches. Check integration with an\nadvanced target before submission; if an update is needed, handle only this\ntask's branch and rerun checks\u002Freview affected by any content change. Do not\nforce a detached checkout, one-commit topology, or history rewrite to manufacture\na clean handoff.\n\nUse [verification](..\u002Fcode-change-verification\u002FSKILL.md) and\n[final review](..\u002Fimplementation-final-review\u002FSKILL.md) before pushing or creating\nor updating a PR. Follow the [release guide](..\u002F..\u002F..\u002F.changeset\u002FREADME.md):\nordinary user-visible SDK changes need a changeset; internal workflow changes\nneed none. Leave version and changelog generation to the release PR.\n\n## Handoff\n\nUse [pr-draft-summary](..\u002Fpr-draft-summary\u002FSKILL.md) for the final description\nand authorized submission\u002Ffollow-up.\n\nInspect status, full diff statistics, and every task commit before staging or\nreporting. Stage only task-owned deliverables; keep local planning and review\nartifacts out of the PR. Preserve hooks and inspect the committed result.\n\nReport branch, base, commit, requested outcome, verification, review status, and\nremaining blockers. When authorized to submit, follow AGENTS.md for current-head\nCI monitoring, scoped fixes, root-channel review requests, and feedback replies.\nFor a requested stack, create each PR against its predecessor and verify that\nrelationship. Do not assume all workflows run on feature-branch PR targets:\ninspect triggers and use an authorized manual CI run on each head when needed.\nNever merge or close the source PR without authorization.\n",{"name":5,"description":7},"main","1f5b2d5d31f70e4766800da41d23aefdf1fbdf04",".agents\u002Fskills\u002Fimplementation-kickoff",22,"https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js","https:\u002F\u002Fguardrails.openai.com","https:\u002F\u002Fgithub.com\u002Fopenai","https:\u002F\u002Fopenai.com\u002F","2026-11-11 05:23:00",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add openai\u002Fopenai-guardrails-js --skill implementation-kickoff",{"tool":42,"command":43},"bunx","bunx skills add openai\u002Fopenai-guardrails-js --skill implementation-kickoff",{"tool":45,"command":46},"pnpm","pnpm dlx skills add openai\u002Fopenai-guardrails-js --skill implementation-kickoff","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fimplementation-kickoff\u002FSKILL.md",true,[50],{"path":51,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":54,"sourceHtmlUrl":55},"agents\u002Fopenai.yaml","REFERENCE","参考文档",125,"https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fimplementation-kickoff\u002Fagents\u002Fopenai.yaml",[57,66,75,85],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":22},"01M6AC5T7FZE6EDYK5PBXQKADB","pr-draft-summary","Draft a Guardrails PR from its complete final diff and carry out authorized CI and review follow-up, including stacked PRs and takeovers.","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpr-draft-summary\u002FSKILL.md",[],[64,20,65],"pr draft summary","review",{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":22},"01M6AC5S48FHHDSWTY69F07P9Z","implementation-strategy","Choose bounded implementation scope and existing owning modules before Guardrails runtime, configuration, client, resource, streaming, Agents, or SDK compatibility changes and feedback fixes.","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fimplementation-strategy\u002FSKILL.md",[],[73,74],"implementation strategy","sdk",{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":22},"01M6AC5NWMTSRH2B62H8Z1QX8G","implementation-final-review","Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure.","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fimplementation-final-review\u002FSKILL.md",[],[65,82,83,20,84],"implementation final review","change","prepare",{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":96,"publishedAt":22},"01M6AC5MB35097EGZ1RPNZZWFK","code-change-verification","Select and run Guardrails verification for code, dependency, test, tooling, documentation, or repository-workflow changes.","https:\u002F\u002Fgithub.com\u002Fopenai\u002Fopenai-guardrails-js\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcode-change-verification\u002FSKILL.md",[91],{"categoryNo":92,"categoryName":93,"taxonomyType":94,"taxonomyTypeDesc":95},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[21,97,83,98,99,100,101,102],"code","code change verification","workflow","run","test","documentation"]