[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75TFH3EDD2XBKD8RNA1GAQ9":3,"skill-same:01M75TFH3EDD2XBKD8RNA1GAQ9":49},{"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":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":31,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M75TFH3EDD2XBKD8RNA1GAQ9","pr-shepherd",null,"Takes a finished eve branch from review to merge. Opens the PR early, runs a fresh-context reviewer subagent, fixes critical findings, addresses, replies to, and resolves review comments, pushes, monitors CI (rerunning flaky jobs or fixing real failures), then squash-merges or enables auto-merge. Follow only when the user explicitly asks to shepherd, babysit, land, or merge a PR, or to \"get a PR green\". Do not start it on your own after finishing an implementation or when only asked to open a PR.","vercel\u002Feve","vercel","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F14985020?v=4","01M75TEEHHNJR9E6CG84ENB3MS","Vercel","01M75TEEJ5BJVTCPRH9YAWSMW4",5902,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpr-shepherd\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: pr-shepherd\ndescription: Takes a finished eve branch from review to merge. Opens the PR early, runs a fresh-context reviewer subagent, fixes critical findings, addresses, replies to, and resolves review comments, pushes, monitors CI (rerunning flaky jobs or fixing real failures), then squash-merges or enables auto-merge. Follow only when the user explicitly asks to shepherd, babysit, land, or merge a PR, or to \"get a PR green\". Do not start it on your own after finishing an implementation or when only asked to open a PR.\nmetadata:\n  internal: true\n---\n\n# PR shepherd\n\nStart after the first implementation is complete and checked locally. Finish\nwhen the PR has no unresolved review threads, no unanswered comments, and green\nCI on the latest head commit, and it is merged or set to auto-merge.\n\nOpen the PR first. Review bots and teammates comment while the rest of the work\nruns, so an early PR gets more feedback in each round.\n\n## 1. Open the PR\n\n1. Make sure the branch is committed with `git commit -s` and that commits are\n   signed (see [`AGENTS.md`](..\u002F..\u002F..\u002FAGENTS.md)). Add a changeset if the change\n   touches the published `eve` package.\n2. Run the fast checks from `AGENTS.md`: `pnpm fmt`, `pnpm lint`,\n   `pnpm typecheck`, plus the narrowest relevant unit tests.\n3. Push the branch, then open the PR. Use the\n   [`gh-pr-description`](..\u002Fgh-pr-description\u002FSKILL.md) skill for the body and\n   pass it with `--body-file`. Open it ready for review, not as a draft, unless\n   the user asks for a draft.\n4. Record the PR number. Every later step uses it.\n\nIf a PR already exists for the branch (`gh pr view --json number,url`), reuse it.\n\n## 2. Fresh-context review\n\nSpawn one reviewer subagent with no conversation history. Give it the PR number\nand the branch, not your reasoning. A fresh reviewer catches what the author\nexplains away. Use a prompt like this one:\n\n> Review PR #\u003Cn> in this repository (`gh pr view \u003Cn>`, `gh pr diff \u003Cn>`). Read\n> `AGENTS.md` first and apply its coding, security, testing, docs, and changeset\n> rules. Read surrounding code as needed to verify behavior. Do not edit files.\n> Report findings ordered by severity (critical, important, minor). For each\n> finding, give the file and line, the concrete failure or risk, and a suggested\n> fix. Say plainly when you find nothing critical.\n\nWhile the reviewer runs, you can already collect PR comments (step 3).\n\nTriage the findings yourself. Verify each one against the code before you act.\n\n- Fix: correctness bugs, races, security or data-loss risks, broken public\n  behavior or docs, missing tests for changed behavior, a missing changeset,\n  and invariant violations.\n- Skip: style nits, speculative refactors, and scope creep. Mention notable\n  skipped items to the user in the final report.\n\n## 3. Address PR comments\n\nCollect all feedback.\n\nUnresolved inline review threads, with the IDs you need to reply and resolve:\n\n```sh\ngh api graphql -F owner='{owner}' -F repo='{repo}' -F number=\u003Cn> -f query='\nquery($owner: String!, $repo: String!, $number: Int!) {\n  repository(owner: $owner, name: $repo) {\n    pullRequest(number: $number) {\n      reviewThreads(first: 100) {\n        nodes {\n          id isResolved isOutdated path line\n          comments(first: 50) { nodes { author { login } body url } }\n        }\n      }\n    }\n  }\n}' --jq '.data.repository.pullRequest.reviewThreads.nodes[] | select(.isResolved | not)'\n```\n\nTop-level comments and review summaries (these cannot be resolved):\n\n```sh\ngh pr view \u003Cn> --json comments,reviews \\\n  --jq '.comments[], .reviews[] | select(.body != \"\") | {author: .author.login, body}'\n```\n\nExpect two review bots:\n\n- `roboeve[bot]` reviews once, on the commit that opened the PR, usually within\n  2–15 minutes. It does not review later pushes.\n- Vercel Agent Review (`vercel[bot]`) runs as a check on every push and can post\n  new threads on later commits. It re-checks its own threads after a fix and\n  resolves them itself, so skip threads it already resolved.\n\nIgnore informational bot posts such as Vercel deployment and bundle analysis\ncomments, unless they report a problem this PR caused.\n\nFor each thread, decide: fix, decline with a reason, or defer to a follow-up.\nApply the same triage bar as step 2. Bot suggestions get the same scrutiny as\nhuman ones; confirm the problem is real before you change code. If a comment\nneeds a product or design decision only the user can make, ask the user.\n\n## 4. Commit, push, reply, resolve\n\n1. Combine the fixes from steps 2 and 3. Rerun the fast checks and the\n   narrowest tests that cover the fixes.\n2. Commit with `git commit -s` and push. Add new commits instead of amending or\n   force-pushing, so reply SHAs stay valid. PRs are squash-merged and `main`\n   does not require branches to be up to date, so do not rebase only to catch\n   up. If `gh pr view \u003Cn> --json mergeable` reports `CONFLICTING`, merge\n   `origin\u002Fmain` into the branch with `git merge --signoff origin\u002Fmain`. Rebase\n   or force-push only when the user asks or to repair signing or DCO.\n3. If the fixes changed behavior that the PR body describes, update the body\n   with `gh-pr-description`.\n4. Reply in each thread after the fix is on the remote. Keep replies short and\n   specific, in the style already used on eve PRs:\n   - `Fixed in \u003Csha>. \u003CWhat changed, in one or two sentences.>`\n   - `Not changing this: \u003Creason grounded in the code or a repo rule>.`\n   - `Agreed, this is a real gap. Tracking it in a follow-up: \u003Cscope>.`\n\n   Reply by thread ID. The REST replies endpoint accepts only a thread's first\n   comment, so it fails on later rounds.\n\n   ```sh\n   gh api graphql -f id='\u003Cthread id>' -f body='Fixed in abc1234. ...' -f query='\n   mutation($id: ID!, $body: String!) { addPullRequestReviewThreadReply(input: { pullRequestReviewThreadId: $id, body: $body }) { comment { url } } }'\n   ```\n\n5. Resolve threads that are fixed and pushed. Also resolve bot threads that\n   you declined or deferred, after you reply. Leave declined or deferred human\n   threads open, so the reviewer can accept the reason before merge, and list\n   them in the final report.\n\n   ```sh\n   gh api graphql -f id='\u003Cthread id>' -f query='\n   mutation($id: ID!) { resolveReviewThread(input: { threadId: $id }) { thread { isResolved } } }'\n   ```\n\n6. Answer top-level comments that ask for a change or a question with\n   `gh pr comment \u003Cn> --body-file \u003Cfile>`.\n\n## 5. Monitor CI\n\nWait for checks on the new head commit. It can take a minute for them to\nregister after a push.\n\n```sh\ngh pr checks \u003Cn> --watch --interval 60\n```\n\nIf your shell cannot block that long, poll `gh pr checks \u003Cn>` on an interval.\nA full round takes about 10–15 minutes, and `test-scenario` and the e2e shards\nfinish last. `gh pr checks \u003Cn> --required` shows only the checks that block\nmerge. Advisory checks such as Bundle Analysis and the eve-code benchmark\nshould still pass, but they come second to required checks.\n\nFor each failure, find the real error before you act. Integration tests log\nmany expected errors, so the first `Error:` lines in a failed log are usually\nnoise. Look for the vitest summary instead:\n\n```sh\ngh run view \u003Crun-id> --log-failed | grep -aE 'Failed Tests|FAIL |Test Files|Unhandled'\n```\n\nIf every test passed but the job still failed, read the end of the log for a\nteardown error, hang, or timeout.\n\nClassify the failure:\n\n- Real: the failure relates to the diff or reproduces locally with the tier\n  config from `AGENTS.md`. Fix it, then go back to step 4.\n- Already failing on `main`: the same job fails the same way on recent `main`\n  runs (`gh run list --branch main --workflow \u003Cfile> --limit 10`). A rerun\n  cannot pass, and the fix does not belong in this PR. Do not rerun it; report\n  it to the user as unrelated.\n- Flaky: the failure does not relate to the diff, and it shows timeouts,\n  network, provider, or deploy errors. Most flakes come from `test-integration`\n  on either OS, `test-scenario`, and `e2e-vercel` or `e2e-postgres` fixture\n  deploys. `e2e-local` uses live models, so it can also vary between runs.\n\nRerun only the failed jobs of a flaky failure. GitHub refuses a rerun while any\njob in that workflow run is still in progress, so wait for the run to finish:\n\n```sh\ngh run rerun \u003Crun-id> --failed\n```\n\nRerun a flaky job at most twice per head commit. If it still fails the same way,\ntreat it as real, or report it to the user with the failing test names. Do not\nrun whole integration, scenario, or e2e tiers locally. Scope local runs to the\nfailing file and test name.\n\n## 6. Loop until done\n\nAfter each push, return to step 3. Teammates may comment at any time, and\nVercel Agent Review reviews each new commit. Stop when all of these are true for\nthe latest head commit:\n\n- All checks are green, including the Vercel Agent Review check, or the only\n  failures are documented as unrelated and the user knows about them.\n- There are no unresolved review threads, except ones that wait on a human\n  decision.\n- No new comments arrived since your last pass.\n\n## 7. Merge\n\nWhen step 6 is done, merge the PR, unless the user asked you not to merge. Check\nthe review state and whether auto-merge is already on:\n\n```sh\ngh pr view \u003Cn> --json reviewDecision,mergeStateStatus,autoMergeRequest,headRefOid\n```\n\nPRs to `main` are squash-merged. GitHub uses the PR title as the commit subject\nand leaves the body blank, so pass the DCO sign-off as the body. Pin the head\ncommit so a push that arrives after your last check cannot merge unseen.\n\n- `APPROVED`, with green required checks: squash-merge now.\n\n  ```sh\n  gh pr merge \u003Cn> --squash --match-head-commit \u003CheadRefOid> \\\n    --body \"Signed-off-by: $(git config user.name) \u003C$(git config user.email)>\"\n  ```\n\n- `REVIEW_REQUIRED`: enable auto-merge, so GitHub merges once a code owner\n  approves. Skip this when `autoMergeRequest` is already set.\n\n  ```sh\n  gh pr merge \u003Cn> --squash --auto --match-head-commit \u003CheadRefOid> \\\n    --body \"Signed-off-by: $(git config user.name) \u003C$(git config user.email)>\"\n  ```\n\n- `CHANGES_REQUESTED`: do not merge or enable auto-merge. Report the review to\n  the user.\n\nDo not pass `--delete-branch`. GitHub deletes the remote branch after the\nmerge, and the local cleanup can fail in a worktree. Confirm the result with\n`gh pr view \u003Cn> --json state,mergedAt,autoMergeRequest`.\n\n## Final report\n\nGive the user the PR URL and a short account of the work:\n\n- Reviewer findings you fixed, and notable ones you skipped and why.\n- Comments you addressed, declined, or deferred, plus threads still open.\n- CI status on the head SHA, flaky reruns with job names, and real failures\n  you fixed.\n- Merge state: merged with the squash commit SHA, auto-merge enabled and waiting\n  on approval, or not merged and why.\n","# PR shepherd\n\nStart after the first implementation is complete and checked locally. Finish\nwhen the PR has no unresolved review threads, no unanswered comments, and green\nCI on the latest head commit, and it is merged or set to auto-merge.\n\nOpen the PR first. Review bots and teammates comment while the rest of the work\nruns, so an early PR gets more feedback in each round.\n\n## 1. Open the PR\n\n1. Make sure the branch is committed with `git commit -s` and that commits are\n   signed (see [`AGENTS.md`](..\u002F..\u002F..\u002FAGENTS.md)). Add a changeset if the change\n   touches the published `eve` package.\n2. Run the fast checks from `AGENTS.md`: `pnpm fmt`, `pnpm lint`,\n   `pnpm typecheck`, plus the narrowest relevant unit tests.\n3. Push the branch, then open the PR. Use the\n   [`gh-pr-description`](..\u002Fgh-pr-description\u002FSKILL.md) skill for the body and\n   pass it with `--body-file`. Open it ready for review, not as a draft, unless\n   the user asks for a draft.\n4. Record the PR number. Every later step uses it.\n\nIf a PR already exists for the branch (`gh pr view --json number,url`), reuse it.\n\n## 2. Fresh-context review\n\nSpawn one reviewer subagent with no conversation history. Give it the PR number\nand the branch, not your reasoning. A fresh reviewer catches what the author\nexplains away. Use a prompt like this one:\n\n> Review PR #\u003Cn> in this repository (`gh pr view \u003Cn>`, `gh pr diff \u003Cn>`). Read\n> `AGENTS.md` first and apply its coding, security, testing, docs, and changeset\n> rules. Read surrounding code as needed to verify behavior. Do not edit files.\n> Report findings ordered by severity (critical, important, minor). For each\n> finding, give the file and line, the concrete failure or risk, and a suggested\n> fix. Say plainly when you find nothing critical.\n\nWhile the reviewer runs, you can already collect PR comments (step 3).\n\nTriage the findings yourself. Verify each one against the code before you act.\n\n- Fix: correctness bugs, races, security or data-loss risks, broken public\n  behavior or docs, missing tests for changed behavior, a missing changeset,\n  and invariant violations.\n- Skip: style nits, speculative refactors, and scope creep. Mention notable\n  skipped items to the user in the final report.\n\n## 3. Address PR comments\n\nCollect all feedback.\n\nUnresolved inline review threads, with the IDs you need to reply and resolve:\n\n```sh\ngh api graphql -F owner='{owner}' -F repo='{repo}' -F number=\u003Cn> -f query='\nquery($owner: String!, $repo: String!, $number: Int!) {\n  repository(owner: $owner, name: $repo) {\n    pullRequest(number: $number) {\n      reviewThreads(first: 100) {\n        nodes {\n          id isResolved isOutdated path line\n          comments(first: 50) { nodes { author { login } body url } }\n        }\n      }\n    }\n  }\n}' --jq '.data.repository.pullRequest.reviewThreads.nodes[] | select(.isResolved | not)'\n```\n\nTop-level comments and review summaries (these cannot be resolved):\n\n```sh\ngh pr view \u003Cn> --json comments,reviews \\\n  --jq '.comments[], .reviews[] | select(.body != \"\") | {author: .author.login, body}'\n```\n\nExpect two review bots:\n\n- `roboeve[bot]` reviews once, on the commit that opened the PR, usually within\n  2–15 minutes. It does not review later pushes.\n- Vercel Agent Review (`vercel[bot]`) runs as a check on every push and can post\n  new threads on later commits. It re-checks its own threads after a fix and\n  resolves them itself, so skip threads it already resolved.\n\nIgnore informational bot posts such as Vercel deployment and bundle analysis\ncomments, unless they report a problem this PR caused.\n\nFor each thread, decide: fix, decline with a reason, or defer to a follow-up.\nApply the same triage bar as step 2. Bot suggestions get the same scrutiny as\nhuman ones; confirm the problem is real before you change code. If a comment\nneeds a product or design decision only the user can make, ask the user.\n\n## 4. Commit, push, reply, resolve\n\n1. Combine the fixes from steps 2 and 3. Rerun the fast checks and the\n   narrowest tests that cover the fixes.\n2. Commit with `git commit -s` and push. Add new commits instead of amending or\n   force-pushing, so reply SHAs stay valid. PRs are squash-merged and `main`\n   does not require branches to be up to date, so do not rebase only to catch\n   up. If `gh pr view \u003Cn> --json mergeable` reports `CONFLICTING`, merge\n   `origin\u002Fmain` into the branch with `git merge --signoff origin\u002Fmain`. Rebase\n   or force-push only when the user asks or to repair signing or DCO.\n3. If the fixes changed behavior that the PR body describes, update the body\n   with `gh-pr-description`.\n4. Reply in each thread after the fix is on the remote. Keep replies short and\n   specific, in the style already used on eve PRs:\n   - `Fixed in \u003Csha>. \u003CWhat changed, in one or two sentences.>`\n   - `Not changing this: \u003Creason grounded in the code or a repo rule>.`\n   - `Agreed, this is a real gap. Tracking it in a follow-up: \u003Cscope>.`\n\n   Reply by thread ID. The REST replies endpoint accepts only a thread's first\n   comment, so it fails on later rounds.\n\n   ```sh\n   gh api graphql -f id='\u003Cthread id>' -f body='Fixed in abc1234. ...' -f query='\n   mutation($id: ID!, $body: String!) { addPullRequestReviewThreadReply(input: { pullRequestReviewThreadId: $id, body: $body }) { comment { url } } }'\n   ```\n\n5. Resolve threads that are fixed and pushed. Also resolve bot threads that\n   you declined or deferred, after you reply. Leave declined or deferred human\n   threads open, so the reviewer can accept the reason before merge, and list\n   them in the final report.\n\n   ```sh\n   gh api graphql -f id='\u003Cthread id>' -f query='\n   mutation($id: ID!) { resolveReviewThread(input: { threadId: $id }) { thread { isResolved } } }'\n   ```\n\n6. Answer top-level comments that ask for a change or a question with\n   `gh pr comment \u003Cn> --body-file \u003Cfile>`.\n\n## 5. Monitor CI\n\nWait for checks on the new head commit. It can take a minute for them to\nregister after a push.\n\n```sh\ngh pr checks \u003Cn> --watch --interval 60\n```\n\nIf your shell cannot block that long, poll `gh pr checks \u003Cn>` on an interval.\nA full round takes about 10–15 minutes, and `test-scenario` and the e2e shards\nfinish last. `gh pr checks \u003Cn> --required` shows only the checks that block\nmerge. Advisory checks such as Bundle Analysis and the eve-code benchmark\nshould still pass, but they come second to required checks.\n\nFor each failure, find the real error before you act. Integration tests log\nmany expected errors, so the first `Error:` lines in a failed log are usually\nnoise. Look for the vitest summary instead:\n\n```sh\ngh run view \u003Crun-id> --log-failed | grep -aE 'Failed Tests|FAIL |Test Files|Unhandled'\n```\n\nIf every test passed but the job still failed, read the end of the log for a\nteardown error, hang, or timeout.\n\nClassify the failure:\n\n- Real: the failure relates to the diff or reproduces locally with the tier\n  config from `AGENTS.md`. Fix it, then go back to step 4.\n- Already failing on `main`: the same job fails the same way on recent `main`\n  runs (`gh run list --branch main --workflow \u003Cfile> --limit 10`). A rerun\n  cannot pass, and the fix does not belong in this PR. Do not rerun it; report\n  it to the user as unrelated.\n- Flaky: the failure does not relate to the diff, and it shows timeouts,\n  network, provider, or deploy errors. Most flakes come from `test-integration`\n  on either OS, `test-scenario`, and `e2e-vercel` or `e2e-postgres` fixture\n  deploys. `e2e-local` uses live models, so it can also vary between runs.\n\nRerun only the failed jobs of a flaky failure. GitHub refuses a rerun while any\njob in that workflow run is still in progress, so wait for the run to finish:\n\n```sh\ngh run rerun \u003Crun-id> --failed\n```\n\nRerun a flaky job at most twice per head commit. If it still fails the same way,\ntreat it as real, or report it to the user with the failing test names. Do not\nrun whole integration, scenario, or e2e tiers locally. Scope local runs to the\nfailing file and test name.\n\n## 6. Loop until done\n\nAfter each push, return to step 3. Teammates may comment at any time, and\nVercel Agent Review reviews each new commit. Stop when all of these are true for\nthe latest head commit:\n\n- All checks are green, including the Vercel Agent Review check, or the only\n  failures are documented as unrelated and the user knows about them.\n- There are no unresolved review threads, except ones that wait on a human\n  decision.\n- No new comments arrived since your last pass.\n\n## 7. Merge\n\nWhen step 6 is done, merge the PR, unless the user asked you not to merge. Check\nthe review state and whether auto-merge is already on:\n\n```sh\ngh pr view \u003Cn> --json reviewDecision,mergeStateStatus,autoMergeRequest,headRefOid\n```\n\nPRs to `main` are squash-merged. GitHub uses the PR title as the commit subject\nand leaves the body blank, so pass the DCO sign-off as the body. Pin the head\ncommit so a push that arrives after your last check cannot merge unseen.\n\n- `APPROVED`, with green required checks: squash-merge now.\n\n  ```sh\n  gh pr merge \u003Cn> --squash --match-head-commit \u003CheadRefOid> \\\n    --body \"Signed-off-by: $(git config user.name) \u003C$(git config user.email)>\"\n  ```\n\n- `REVIEW_REQUIRED`: enable auto-merge, so GitHub merges once a code owner\n  approves. Skip this when `autoMergeRequest` is already set.\n\n  ```sh\n  gh pr merge \u003Cn> --squash --auto --match-head-commit \u003CheadRefOid> \\\n    --body \"Signed-off-by: $(git config user.name) \u003C$(git config user.email)>\"\n  ```\n\n- `CHANGES_REQUESTED`: do not merge or enable auto-merge. Report the review to\n  the user.\n\nDo not pass `--delete-branch`. GitHub deletes the remote branch after the\nmerge, and the local cleanup can fail in a worktree. Confirm the result with\n`gh pr view \u003Cn> --json state,mergedAt,autoMergeRequest`.\n\n## Final report\n\nGive the user the PR URL and a short account of the work:\n\n- Reviewer findings you fixed, and notable ones you skipped and why.\n- Comments you addressed, declined, or deferred, plus threads still open.\n- CI status on the head SHA, flaky reruns with job names, and real failures\n  you fixed.\n- Merge state: merged with the squash commit SHA, auto-merge enabled and waiting\n  on approval, or not merged and why.\n",{"name":5,"metadata":24,"description":7},{"internal":25},true,"main","a318dcf054343b7a9007dd4e4367ac19374a6809",".agents\u002Fskills\u002Fpr-shepherd",656,"https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve","https:\u002F\u002Feve.dev","https:\u002F\u002Fgithub.com\u002Fvercel","https:\u002F\u002Fvercel.com","Agentic infrastructure for every app and agent.","2026-11-15 05:08:34",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":25},[38,41,44],{"tool":39,"command":40},"npx","npx skills add vercel\u002Feve --skill pr-shepherd",{"tool":42,"command":43},"bunx","bunx skills add vercel\u002Feve --skill pr-shepherd",{"tool":45,"command":46},"pnpm","pnpm dlx skills add vercel\u002Feve --skill pr-shepherd","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fpr-shepherd\u002FSKILL.md",[],[50,58,64,71,78,85],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":57},"01M7DYJETPKENT6T9THCB7WKFD","writing-quality","The review rubric: writing-quality rules and reference lists for judging a content draft, covering AI-sounding phrasing, bloated wording, structure, and web-content specs. Load at the start of every review.","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002Fapps\u002Ftemplates\u002Fsanity-copilot-eve-template\u002Fagent\u002Fsubagents\u002Freviewer\u002Fskills\u002Fwriting-quality\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":59,"name":52,"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":57},"01M7DYJAZM6Y5VNF2PCM01N145","Writing-quality guardrails for any prose the agent drafts or edits: blog posts, page copy, Notion drafts, release notes, marketing text. Use this skill whenever writing or revising content meant for humans to read, to keep the prose natural, plain, and free of AI-sounding phrasing. Not needed for code, queries, or structural CMS work.","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002Fapps\u002Ftemplates\u002Fsanity-copilot-eve-template\u002Fagent\u002Fskills\u002Fwriting-quality\u002FSKILL.md",[],[],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":57},"01M7DYJ40G4YADA5P8YAEQ7W7F","seo-aeo-best-practices","SEO and AEO best practices for metadata, Open Graph, sitemaps, robots.txt, hreflang, JSON-LD structured data, EEAT, and content optimized for search engines and AI answer surfaces. Use this skill when implementing page SEO, technical SEO, schema markup, international SEO, AI-overview readiness, or improving content for Google, ChatGPT, Perplexity, and similar assistants.","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002Fapps\u002Ftemplates\u002Fsanity-copilot-eve-template\u002Fagent\u002Fskills\u002Fseo-aeo-best-practices\u002FSKILL.md",[],[],{"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":57},"01M7DYHX5JXJFHF2DZV64VFC13","sanity-best-practices","Sanity development best practices for schema design, GROQ queries, TypeGen, Visual Editing, images, Portable Text, Studio structure, localization, migrations, Sanity Functions, Blueprints, and framework integrations such as Next.js, Nuxt, Astro, Remix, SvelteKit, Angular, Hydrogen, and the App SDK. Use this skill whenever working with Sanity schemas, defineType or defineField, GROQ or defineQuery, content modeling, Presentation or preview setups, Sanity-powered frontend integrations, Sanity Functions, documentEventHandler, defineDocumentFunction, defineMediaLibraryAssetFunction, @sanity\u002Ffunctions, @sanity\u002Fblueprints, sanity.blueprint.ts, event-driven content automation, or when reviewing and fixing a Sanity codebase.","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002Fapps\u002Ftemplates\u002Fsanity-copilot-eve-template\u002Fagent\u002Fskills\u002Fsanity-best-practices\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":57},"01M7DYHQG96APC7EV3FANC90WX","portable-text-serialization","Render and serialize Portable Text to React, Svelte, Vue, Astro, HTML, Markdown, and plain text. Use when implementing Portable Text rendering in any frontend framework, building custom serializers for non-standard block types, converting Portable Text to HTML strings server-side, converting Portable Text to Markdown, extracting plain text from Portable Text, or troubleshooting rendering issues with marks, blocks, lists, or custom types.","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002Fapps\u002Ftemplates\u002Fsanity-copilot-eve-template\u002Fagent\u002Fskills\u002Fportable-text-serialization\u002FSKILL.md",[],[],{"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":91,"publishedAt":57},"01M7DYHJJ89WGRXXVWJAQNQW9M","portable-text-conversion","Convert HTML and Markdown content into Portable Text blocks for Sanity. Use when migrating content from legacy CMSs, importing HTML or Markdown into Sanity, building content pipelines that ingest external content, converting rich text between formats, or programmatically creating Portable Text documents. Covers @portabletext\u002Fmarkdown (markdownToPortableText), @portabletext\u002Fblock-tools (htmlToBlocks), custom deserializers, and the Portable Text specification for manual block construction.","https:\u002F\u002Fgithub.com\u002Fvercel\u002Feve\u002Fblob\u002Fmain\u002Fapps\u002Ftemplates\u002Fsanity-copilot-eve-template\u002Fagent\u002Fskills\u002Fportable-text-conversion\u002FSKILL.md",[],[]]