[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7DM2146TCNYXVVS2QXF08GS":3,"skill-same:01M7DM2146TCNYXVVS2QXF08GS":46},{"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":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M7DM2146TCNYXVVS2QXF08GS","pr-review",null,"Review a pull request on the owner's behalf and fix what's safe to fix. Runs the deterministic gates first, then correctness\u002Fsecurity, design-system, accessibility and analytics review passes, verifies each blocking finding, applies safe fixes, re-checks, and ends with a plain-English verdict. Use when a PR is opened, when asked to review a PR, a branch or a pasted diff, or before merging anything.","foxinthehenhouse\u002Fapp-in-a-box","foxinthehenhouse","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F144333692?v=4","01M7DKXRK1CCGY9F45FYGWVD29","Kyle Fox","01M7DKXRKNG8A1ADJFXVH15RXT",0,"MIT","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fpr-review\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: pr-review\ndescription: Review a pull request on the owner's behalf and fix what's safe to fix. Runs the deterministic gates first, then correctness\u002Fsecurity, design-system, accessibility and analytics review passes, verifies each blocking finding, applies safe fixes, re-checks, and ends with a plain-English verdict. Use when a PR is opened, when asked to review a PR, a branch or a pasted diff, or before merging anything.\nallowed-tools: \"Bash(gh:*), Bash(git:*), Bash(scripts\u002Fdev-venv.sh:*), Bash(npm:*), Bash(cd:*), Read, Glob, Grep, Edit, Write\"\n---\n\n# PR review: the gate before main\n\nAssume the owner doesn't read diffs. This review is the only thing standing between a\nchange and `main`, so be the careful engineer they don't have.\n\n> **Thorough mode (Claude Code only, opt-in, ~5–10× the tokens):** if the owner asks\n> for the workflow review, run the `pr-review` Workflow (`.claude\u002Fworkflows\u002Fpr-review.js`)\n> instead of steps 1–4. It fans out the same passes, has two skeptics verify each\n> finding and loops the fixes, then hands you its result for step 5. This skill is\n> the default route, and the only one in Codex.\n\n## 0. Target\n\nAn explicit PR number, URL or branch; otherwise the PR for the current branch\n(`gh pr view --json number,title,headRefName,baseRefName,isDraft,files,url`). Diff =\n`origin\u002F\u003Cbase>...HEAD`. Drafts get reviewed but never merged.\n\n**Handed an inline diff** (pasted into the prompt, with no PR and no checkout)? Review\nit as the PR: the same passes, the same verdict shape. The gates in step 1 can't run,\nso name that in the verdict (\"What's left: gates not run, no checkout\") instead of\nskipping it silently, and judge the tests by reading them: would each one go red if\nthe code were wrong? No PR and no diff means you stop and say so.\n\n## 1. Deterministic gates (cheap, fail fast)\n\nRun only what the changed files touch:\n- Backend or `*.py`: `scripts\u002Fdev-venv.sh python -m pytest -q`, then\n  `scripts\u002Fdev-venv.sh ruff check backend tests`.\n- Mobile: `cd mobile && npm run gates`.\n- Migrations: read each new file against `.agents\u002Frules\u002Fdb-migrations.md`.\n- A v1 response model changed: confirm `mobile\u002Flib\u002Fapi.ts` `*Wire` types match.\n\nA red gate is finding #1. Fix it if the fix is mechanical; otherwise report it.\n\n## 2. Review passes\n\nRun these as parallel subagents if your agent supports them; otherwise sequentially.\nEach pass returns findings as `severity (blocker|major|minor) · file:line · what\nbreaks · for whom`.\n\n1. **Correctness + security:** logic bugs, unhandled errors, missing `user_id`\n   scoping (IDOR), secrets in code, trusting ids from request bodies, in-process\n   state, breaking wire changes (`.agents\u002Frules\u002Fapi-contract.md`).\n2. **Design system + accessibility:** hex literals, raw sizes, non-ink text colours,\n   missing `accessibilityRole`\u002Flabel, tap targets \u003C 48, meaning-by-colour-only, and\n   the `docs\u002Fdesign\u002FTASTE.md` rubric on any new or changed screen (one primary\n   action, honest states, restraint, copy that is a verb and not a form label).\n3. **Analytics + env wiring:** new screen without a view event, mutation without\n   success+failure events, new `EXPO_PUBLIC_*` or backend env var not wired\n   (`.agents\u002Frules\u002Fenv-var-wiring.md`).\n4. **Context docs:** does `AGENTS.md` still describe the code after this PR? The map\n   row for anything added, and any rule, product line or command the diff made\n   untrue (the lint only checks the map and markers; the prose is yours to read).\n5. **Tests:** is the new behaviour tested? Would the test fail if the code were\n   wrong? (Negative-control one important test mentally: delete the key line, and\n   does anything go red?)\n\n## 3. Verify before you believe\n\nFor every blocker\u002Fmajor: re-read the code path and try to disprove the finding.\nDrop what doesn't survive. Reviewers hallucinate; the verdict must not.\n\n## 4. Fix what's safe\n\nSafe: mechanical lint\u002Fformat, missing labels\u002FtestIDs, missing analytics call on a new\nscreen, an obvious missing `.eq(\"user_id\", ...)` with a test proving it. **Not** safe:\nanything changing product behaviour, schema or wire shape; ask instead. Never weaken\nor delete a test to go green. Commit fixes to the PR branch\n(`fix(review): \u003Cwhat>`), re-run the gates, and loop until green or 3 rounds.\n\n## 4b. Chair ruling (risky PRs only)\n\nRisky = touches auth\u002FRLS, a migration, secrets\u002Fenv\u002FCI\u002Fbuild config, deletes files, or\nis larger than ~400 net lines. For those, spawn the `chair` subagent once, with the\ngate output, the verified findings and your classification (not the PR description:\nit reads the diff first). In Codex, run the chair's instructions from\n`.agents\u002Fagents\u002Fchair.md` as a final pass. An ESCALATE ruling means no merge. MERGE\non a risky PR still goes to the owner, because the chair can veto but can't authorise.\n\n## 5. Verdict (plain English, top of your reply and as a PR comment)\n\n```\nVerdict: ✅ Safe to merge | ⚠️ Merge after owner decision | ❌ Not ready\nWhat this PR does: \u003Cone sentence a non-engineer understands>\nWhat I fixed: \u003Cbullets or \"nothing\">\nWhat's left: \u003Cblockers with file:line, or \"nothing\">\nRisk: low | medium | high, because \u003Creason>\nChair: \u003CMERGE\u002FESCALATE + its three reasons, or \"not needed (low risk)\">\n\u003C!-- appbox-verdict sha=\u003Cfull head sha you reviewed> result=safe|owner|not-ready -->\n```\n\nThe last line is for the `land` skill: it is how the loop knows THIS commit was\nreviewed (a later push needs a new verdict). `safe` = ✅, `owner` = ⚠️, `not-ready` = ❌.\nGet the sha with `gh pr view --json headRefOid -q .headRefOid`.\n\nMerge without asking only if `appbox.yaml` → `policy.auto_merge_low_risk` is `true`\n(default `false`; only the owner flips it) or the owner said so in this conversation,\nAND the verdict is ✅, the risk is low and CI is green. That is the same predicate\n`land` takes as `--merge-ok`. Otherwise the owner merges. Squash merge, delete the\nbranch. To take a PR the rest of the way (CI, threads, re-review after each push,\nmerge), hand it to the `land` skill.\n\n## Ask the owner\n\nFollow `.agents\u002Frules\u002Fproduct-judgement.md`: ask with a structured question (recommended option first), never decide these silently. In this skill that means: Findings that are product calls, not bugs: changed copy or flow, scope beyond the ticket, a new data field, a known limitation. Interactive: ask the owner before approving. In CI (review-only): post them as ⚖️ questions in the verdict comment and don't count them as blockers or approvals.\n","# PR review: the gate before main\n\nAssume the owner doesn't read diffs. This review is the only thing standing between a\nchange and `main`, so be the careful engineer they don't have.\n\n> **Thorough mode (Claude Code only, opt-in, ~5–10× the tokens):** if the owner asks\n> for the workflow review, run the `pr-review` Workflow (`.claude\u002Fworkflows\u002Fpr-review.js`)\n> instead of steps 1–4. It fans out the same passes, has two skeptics verify each\n> finding and loops the fixes, then hands you its result for step 5. This skill is\n> the default route, and the only one in Codex.\n\n## 0. Target\n\nAn explicit PR number, URL or branch; otherwise the PR for the current branch\n(`gh pr view --json number,title,headRefName,baseRefName,isDraft,files,url`). Diff =\n`origin\u002F\u003Cbase>...HEAD`. Drafts get reviewed but never merged.\n\n**Handed an inline diff** (pasted into the prompt, with no PR and no checkout)? Review\nit as the PR: the same passes, the same verdict shape. The gates in step 1 can't run,\nso name that in the verdict (\"What's left: gates not run, no checkout\") instead of\nskipping it silently, and judge the tests by reading them: would each one go red if\nthe code were wrong? No PR and no diff means you stop and say so.\n\n## 1. Deterministic gates (cheap, fail fast)\n\nRun only what the changed files touch:\n- Backend or `*.py`: `scripts\u002Fdev-venv.sh python -m pytest -q`, then\n  `scripts\u002Fdev-venv.sh ruff check backend tests`.\n- Mobile: `cd mobile && npm run gates`.\n- Migrations: read each new file against `.agents\u002Frules\u002Fdb-migrations.md`.\n- A v1 response model changed: confirm `mobile\u002Flib\u002Fapi.ts` `*Wire` types match.\n\nA red gate is finding #1. Fix it if the fix is mechanical; otherwise report it.\n\n## 2. Review passes\n\nRun these as parallel subagents if your agent supports them; otherwise sequentially.\nEach pass returns findings as `severity (blocker|major|minor) · file:line · what\nbreaks · for whom`.\n\n1. **Correctness + security:** logic bugs, unhandled errors, missing `user_id`\n   scoping (IDOR), secrets in code, trusting ids from request bodies, in-process\n   state, breaking wire changes (`.agents\u002Frules\u002Fapi-contract.md`).\n2. **Design system + accessibility:** hex literals, raw sizes, non-ink text colours,\n   missing `accessibilityRole`\u002Flabel, tap targets \u003C 48, meaning-by-colour-only, and\n   the `docs\u002Fdesign\u002FTASTE.md` rubric on any new or changed screen (one primary\n   action, honest states, restraint, copy that is a verb and not a form label).\n3. **Analytics + env wiring:** new screen without a view event, mutation without\n   success+failure events, new `EXPO_PUBLIC_*` or backend env var not wired\n   (`.agents\u002Frules\u002Fenv-var-wiring.md`).\n4. **Context docs:** does `AGENTS.md` still describe the code after this PR? The map\n   row for anything added, and any rule, product line or command the diff made\n   untrue (the lint only checks the map and markers; the prose is yours to read).\n5. **Tests:** is the new behaviour tested? Would the test fail if the code were\n   wrong? (Negative-control one important test mentally: delete the key line, and\n   does anything go red?)\n\n## 3. Verify before you believe\n\nFor every blocker\u002Fmajor: re-read the code path and try to disprove the finding.\nDrop what doesn't survive. Reviewers hallucinate; the verdict must not.\n\n## 4. Fix what's safe\n\nSafe: mechanical lint\u002Fformat, missing labels\u002FtestIDs, missing analytics call on a new\nscreen, an obvious missing `.eq(\"user_id\", ...)` with a test proving it. **Not** safe:\nanything changing product behaviour, schema or wire shape; ask instead. Never weaken\nor delete a test to go green. Commit fixes to the PR branch\n(`fix(review): \u003Cwhat>`), re-run the gates, and loop until green or 3 rounds.\n\n## 4b. Chair ruling (risky PRs only)\n\nRisky = touches auth\u002FRLS, a migration, secrets\u002Fenv\u002FCI\u002Fbuild config, deletes files, or\nis larger than ~400 net lines. For those, spawn the `chair` subagent once, with the\ngate output, the verified findings and your classification (not the PR description:\nit reads the diff first). In Codex, run the chair's instructions from\n`.agents\u002Fagents\u002Fchair.md` as a final pass. An ESCALATE ruling means no merge. MERGE\non a risky PR still goes to the owner, because the chair can veto but can't authorise.\n\n## 5. Verdict (plain English, top of your reply and as a PR comment)\n\n```\nVerdict: ✅ Safe to merge | ⚠️ Merge after owner decision | ❌ Not ready\nWhat this PR does: \u003Cone sentence a non-engineer understands>\nWhat I fixed: \u003Cbullets or \"nothing\">\nWhat's left: \u003Cblockers with file:line, or \"nothing\">\nRisk: low | medium | high, because \u003Creason>\nChair: \u003CMERGE\u002FESCALATE + its three reasons, or \"not needed (low risk)\">\n\u003C!-- appbox-verdict sha=\u003Cfull head sha you reviewed> result=safe|owner|not-ready -->\n```\n\nThe last line is for the `land` skill: it is how the loop knows THIS commit was\nreviewed (a later push needs a new verdict). `safe` = ✅, `owner` = ⚠️, `not-ready` = ❌.\nGet the sha with `gh pr view --json headRefOid -q .headRefOid`.\n\nMerge without asking only if `appbox.yaml` → `policy.auto_merge_low_risk` is `true`\n(default `false`; only the owner flips it) or the owner said so in this conversation,\nAND the verdict is ✅, the risk is low and CI is green. That is the same predicate\n`land` takes as `--merge-ok`. Otherwise the owner merges. Squash merge, delete the\nbranch. To take a PR the rest of the way (CI, threads, re-review after each push,\nmerge), hand it to the `land` skill.\n\n## Ask the owner\n\nFollow `.agents\u002Frules\u002Fproduct-judgement.md`: ask with a structured question (recommended option first), never decide these silently. In this skill that means: Findings that are product calls, not bugs: changed copy or flow, scope beyond the ticket, a new data field, a known limitation. Interactive: ask the owner before approving. In CI (review-only): post them as ⚖️ questions in the verdict comment and don't count them as blockers or approvals.\n",{"name":5,"description":7,"allowed-tools":24},"Bash(gh:*), Bash(git:*), Bash(scripts\u002Fdev-venv.sh:*), Bash(npm:*), Bash(cd:*), Read, Glob, Grep, Edit, Write","main","82b5819f074e18b5d31db7a122e1f0ae5f5abbf7","plugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fpr-review","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse","Not a developer","2026-10-06 17:24:46",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add foxinthehenhouse\u002Fapp-in-a-box --skill pr-review",{"tool":38,"command":39},"bunx","bunx skills add foxinthehenhouse\u002Fapp-in-a-box --skill pr-review",{"tool":41,"command":42},"pnpm","pnpm dlx skills add foxinthehenhouse\u002Fapp-in-a-box --skill pr-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fpr-review\u002FSKILL.md",true,[],[47,54,61,68,75,82],{"skillNo":48,"name":49,"nameZh":6,"description":50,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":19},"01M7DM2EPAFBTFNV5EW5GHKD5Q","ship","Release checklist for getting a build to testers or the stores. Runs the risk gate first (refuses while a risk item is open) and walks the owner through the compliance checklist. Covers the version bump, EAS build and submit, OTA updates and when they're safe, changelog, store listing, privacy nutrition label \u002F data safety form, the Accessibility Nutrition Labels, and the account-deletion requirement. Guidance-level; it never handles signing secrets. Use when asked to ship, release, submit to TestFlight \u002F App Store \u002F Play, or push an OTA update.","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fship\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7DM2AGRJ822B3C8NQCDK3JG","routines","Put the project's recurring rituals on a schedule, with the owner's consent. Weekly north-star report, monthly market watch, dependency triage, Supabase advisor sweep, reflect and harness-optimize. Uses Claude Code Routines (cloud cron) where they fit, and documents the Codex \u002F local cron equivalent. Use when asked to automate, schedule or \"keep the project moving on its own\", or after setup finishes.","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Froutines\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M7DM27C3K1Y7FM0NZSBRSVEN","reflect","Consolidate what recent sessions learned into the project's long-term memory (.agents\u002Fmemory). Use at the end of a substantial session, after a correction from the user, after a painful debugging session, when the session-start healthcheck says reflect is overdue or a pending-reflection seed is waiting, or when MEMORY.md exceeds its budget.","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Freflect\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7DM1X6QP3QAT47ET852S0P8","north-star-report","Weekly product-analytics and reliability readout. Pulls the funnel of the brief's 5 north-star events from PostHog, finds the biggest drop-offs and errors, checks each SLO in docs\u002Fslo.yaml against its error budget, compares with last week, and proposes (then, with a yes, files) tickets that target them. Use weekly, when the session-start line says it's overdue, when the backlog is empty, or when asked \"how is the app doing?\".","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fnorth-star-report\u002FSKILL.md",[],[],{"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":19},"01M7DM1QX8F3D95JSTEHC0P9QA","next","Decide the single next best thing to work on, with two alternates. Reads setup progress, work in progress, open PRs and failing CI, the backlog (GitHub Issues or Linear), overdue harness rituals, product decisions parked until now, and north-star analytics if PostHog is connected. Use when someone asks \"what next?\", \"what should I work on?\", at the start of a session, or after a PR merges.","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fnext\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M7DM1NGHSRAY1NJVDWE1TG9Q","new-worktree","Create an isolated git worktree + branch off origin\u002Fmain for one unit of work, so parallel agents or sessions never share a checkout. Use when starting any ticket, or when another session may be working in the main checkout.","https:\u002F\u002Fgithub.com\u002Ffoxinthehenhouse\u002Fapp-in-a-box\u002Fblob\u002Fmain\u002Fplugins\u002Fapp-in-a-box\u002Ftemplate\u002F.agents\u002Fskills\u002Fnew-worktree\u002FSKILL.md",[],[]]