[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79PQ5SQDEMDZ2GACSR64CQV":3,"skill-same:01M79PQ5SQDEMDZ2GACSR64CQV":90},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M79PQ5SQDEMDZ2GACSR64CQV","visual-verification",null,"Capture, inspect, and validate rendered local app UI with headless Chrome screenshots. Use when Codex needs to verify frontend changes, compare light\u002Fdark or responsive states, inspect authenticated local pages, capture hover\u002Fdropdown\u002Ffocus states, diagnose visual regressions, or create a repo-local screenshot\u002Fauth workflow for agents.","nick-neely\u002Fagent-skills","nick-neely","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F49537823?v=4","01M79PN6YVPE30QN6V8RFVE3HX","Nick Neely","01M79PN7002TSYJ8RTQFBT3SDM",0,"MIT","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: visual-verification\ndescription: Capture, inspect, and validate rendered local app UI with headless Chrome screenshots. Use when Codex needs to verify frontend changes, compare light\u002Fdark or responsive states, inspect authenticated local pages, capture hover\u002Fdropdown\u002Ffocus states, diagnose visual regressions, or create a repo-local screenshot\u002Fauth workflow for agents.\n---\n\n# Visual Verification\n\nUse this skill to turn UI claims into screenshot evidence. The goal is a local-agent workflow: discover or create repo config, capture the rendered state, inspect the PNG, and iterate until the UI is actually verified.\n\nIn commands below, replace `\u003Cskill-root>` with the installed directory containing this `SKILL.md`.\n\n## First Use\n\n1. Look for `.agents\u002Fvisual-verification.json`.\n2. If missing, run:\n\n```bash\nnode \"\u003Cskill-root>\u002Fscripts\u002Fdetect-project.mjs\"\nnode \"\u003Cskill-root>\u002Fscripts\u002Finit-config.mjs\"\n```\n\n3. Run browser setup checks when Chrome\u002Fheadless-shell availability is unknown:\n\n```bash\nnode \"\u003Cskill-root>\u002Fscripts\u002Fensure-browser.mjs\" --smoke\n```\n\nUse `--install --smoke` if no browser is found.\n4. Read `references\u002Ffirst-use.md` when config is missing, stale, or the repo needs auth setup.\n5. Prefer existing repo commands over generated helpers, but do not assume a command name.\n6. When protected pages need auth, read `references\u002Fauth-adapters.md` before adding or changing auth setup.\n\nThe user does not need to hand-author the config. Agents may create it automatically after inspecting the repo.\n\n## Screenshot Workflow\n\n1. Reuse an existing dev server when one is running; start the configured dev server only when needed.\n2. Run the screenshot script from the repo root:\n\n```bash\nnode \"\u003Cskill-root>\u002Fscripts\u002Fscreenshot.mjs\" \u003Curl> [options]\n```\n\n3. For UI work, capture the relevant states rather than one resting viewport:\n   - light and dark modes when the app supports themes\n   - desktop and mobile\u002Ftablet when layout changes are relevant\n   - hover, dropdown, focus, open modal, and scrolled states when those are the changed surfaces\n4. Open\u002Fread the PNG before concluding. A successful command is not enough.\n5. If the command fails on a selector, expectation, console error, or network error, treat that as a useful verification failure and fix the underlying issue or the selector.\n\nRead `references\u002Fscreenshot-recipes.md` for command examples.\n\n## Auth Boundary\n\nThe skill owns browser control and visual verification. The repo owns auth mutation.\n\nSupported auth modes in config:\n\n- `none`: public pages only.\n- `storage-state`: use an existing Playwright storage state; never refresh it.\n- `command`: run a repo-owned command that writes a Playwright storage state.\n- `custom-cookie`: a repo-owned command creates cookie\u002Fstorage-state data.\n- `playwright-login`: a repo-owned Playwright script logs in and writes storage state.\n\nFor Better Auth, keep setup local-only and fail closed: refuse remote DB URLs, require `BETTER_AUTH_SECRET`, create an obvious agent user, grant only local verification permissions, and write `.auth\u002Fagent-admin-storage-state.json` with restrictive file permissions.\n\n## Scripts\n\n- `scripts\u002Fscreenshot.mjs`: plain-Node CDP screenshot runner.\n- `scripts\u002Fdetect-project.mjs`: inspect package scripts, routes, and auth signals.\n- `scripts\u002Finit-config.mjs`: write `.agents\u002Fvisual-verification.json`.\n- `scripts\u002Fensure-browser.mjs`: verify or install the Chrome\u002Fheadless-shell runtime.\n- `scripts\u002Fensure-auth.mjs`: refresh configured auth storage state without capturing.\n\n## References\n\n- `references\u002Ffirst-use.md`: config discovery and initialization flow.\n- `references\u002Fscreenshot-recipes.md`: common screenshot commands.\n- `references\u002Fauth-adapters.md`: auth adapter patterns and safety rules.\n- `references\u002Ftroubleshooting.md`: known failure modes.\n","# Visual Verification\n\nUse this skill to turn UI claims into screenshot evidence. The goal is a local-agent workflow: discover or create repo config, capture the rendered state, inspect the PNG, and iterate until the UI is actually verified.\n\nIn commands below, replace `\u003Cskill-root>` with the installed directory containing this `SKILL.md`.\n\n## First Use\n\n1. Look for `.agents\u002Fvisual-verification.json`.\n2. If missing, run:\n\n```bash\nnode \"\u003Cskill-root>\u002Fscripts\u002Fdetect-project.mjs\"\nnode \"\u003Cskill-root>\u002Fscripts\u002Finit-config.mjs\"\n```\n\n3. Run browser setup checks when Chrome\u002Fheadless-shell availability is unknown:\n\n```bash\nnode \"\u003Cskill-root>\u002Fscripts\u002Fensure-browser.mjs\" --smoke\n```\n\nUse `--install --smoke` if no browser is found.\n4. Read `references\u002Ffirst-use.md` when config is missing, stale, or the repo needs auth setup.\n5. Prefer existing repo commands over generated helpers, but do not assume a command name.\n6. When protected pages need auth, read `references\u002Fauth-adapters.md` before adding or changing auth setup.\n\nThe user does not need to hand-author the config. Agents may create it automatically after inspecting the repo.\n\n## Screenshot Workflow\n\n1. Reuse an existing dev server when one is running; start the configured dev server only when needed.\n2. Run the screenshot script from the repo root:\n\n```bash\nnode \"\u003Cskill-root>\u002Fscripts\u002Fscreenshot.mjs\" \u003Curl> [options]\n```\n\n3. For UI work, capture the relevant states rather than one resting viewport:\n   - light and dark modes when the app supports themes\n   - desktop and mobile\u002Ftablet when layout changes are relevant\n   - hover, dropdown, focus, open modal, and scrolled states when those are the changed surfaces\n4. Open\u002Fread the PNG before concluding. A successful command is not enough.\n5. If the command fails on a selector, expectation, console error, or network error, treat that as a useful verification failure and fix the underlying issue or the selector.\n\nRead `references\u002Fscreenshot-recipes.md` for command examples.\n\n## Auth Boundary\n\nThe skill owns browser control and visual verification. The repo owns auth mutation.\n\nSupported auth modes in config:\n\n- `none`: public pages only.\n- `storage-state`: use an existing Playwright storage state; never refresh it.\n- `command`: run a repo-owned command that writes a Playwright storage state.\n- `custom-cookie`: a repo-owned command creates cookie\u002Fstorage-state data.\n- `playwright-login`: a repo-owned Playwright script logs in and writes storage state.\n\nFor Better Auth, keep setup local-only and fail closed: refuse remote DB URLs, require `BETTER_AUTH_SECRET`, create an obvious agent user, grant only local verification permissions, and write `.auth\u002Fagent-admin-storage-state.json` with restrictive file permissions.\n\n## Scripts\n\n- `scripts\u002Fscreenshot.mjs`: plain-Node CDP screenshot runner.\n- `scripts\u002Fdetect-project.mjs`: inspect package scripts, routes, and auth signals.\n- `scripts\u002Finit-config.mjs`: write `.agents\u002Fvisual-verification.json`.\n- `scripts\u002Fensure-browser.mjs`: verify or install the Chrome\u002Fheadless-shell runtime.\n- `scripts\u002Fensure-auth.mjs`: refresh configured auth storage state without capturing.\n\n## References\n\n- `references\u002Ffirst-use.md`: config discovery and initialization flow.\n- `references\u002Fscreenshot-recipes.md`: common screenshot commands.\n- `references\u002Fauth-adapters.md`: auth adapter patterns and safety rules.\n- `references\u002Ftroubleshooting.md`: known failure modes.\n",{"name":5,"description":7},"main","a79415595d32d9de44410555b992c907d2acd39b","skills\u002Fvisual-verification","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills","https:\u002F\u002Fgithub.com\u002Fnick-neely","nickneely.dev","Building web apps, breaking them, and fixing them again. React, Next.js, TypeScript, C#.","2026-11-14 04:25:18",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add nick-neely\u002Fagent-skills --skill visual-verification",{"tool":38,"command":39},"bunx","bunx skills add nick-neely\u002Fagent-skills --skill visual-verification",{"tool":41,"command":42},"pnpm","pnpm dlx skills add nick-neely\u002Fagent-skills --skill visual-verification","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002FSKILL.md",true,[46,52,56,60,64,68,74,78,82,86],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"agents\u002Fopenai.yaml","REFERENCE","参考文档",215,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Fagents\u002Fopenai.yaml",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fauth-adapters.md",3238,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Freferences\u002Fauth-adapters.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ffirst-use.md",2486,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Freferences\u002Ffirst-use.md",{"path":61,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fscreenshot-recipes.md",2138,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Freferences\u002Fscreenshot-recipes.md",{"path":65,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Ftroubleshooting.md",2046,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Freferences\u002Ftroubleshooting.md",{"path":69,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":72,"sourceHtmlUrl":73},"scripts\u002Fdetect-project.mjs","SCRIPT","脚本",7168,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Fscripts\u002Fdetect-project.mjs",{"path":75,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":76,"sourceHtmlUrl":77},"scripts\u002Fensure-auth.mjs",2800,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Fscripts\u002Fensure-auth.mjs",{"path":79,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":80,"sourceHtmlUrl":81},"scripts\u002Fensure-browser.mjs",4604,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Fscripts\u002Fensure-browser.mjs",{"path":83,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":84,"sourceHtmlUrl":85},"scripts\u002Finit-config.mjs",3204,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Fscripts\u002Finit-config.mjs",{"path":87,"resourceType":70,"resourceTypeDesc":71,"sizeBytes":88,"sourceHtmlUrl":89},"scripts\u002Fscreenshot.mjs",37408,"https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fvisual-verification\u002Fscripts\u002Fscreenshot.mjs",[91,98,105,112,117,122],{"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":19},"01M7BECTZNYM56SQ1M304TC6JC","flow-walk","Fresh-eyes walk of one user flow in the running app, judging ease: friction, copy weight, confusing hierarchy and what can be removed. Use before code review when a change alters what a user sees or decides, and when asked to \"walk this flow\", \"try this as a user\" or get \"fresh eyes\" on a screen. The builder hands one user goal to a tester who knows nothing about the change, then sorts the report. It takes no acceptance criteria and keeps no run folder, only the report; checking acceptance criteria against requirements with retained evidence is the uat skill's job, and correctness belongs to tests and code review.","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fflow-walk\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M79PRF9536B79K1KDH3EC2ES","wrong","Wrong name fixture.","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fwrong-name\u002Fskills\u002Fright\u002FSKILL.md",[],[],{"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":19},"01M79PRE3HN1MB02EJ5STBKMET","sample","Sample skill for a fixture.","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Funknown-frontmatter-key\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":113,"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":114,"categories":115,"keywords":116,"publishedAt":19},"01M79PRD25SHAJTDB4JX87REEW","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Funexpected-skill-subdirectory\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":118,"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":119,"categories":120,"keywords":121,"publishedAt":19},"01M79PRBYYVMJ7KXZV11AQR584","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Freadme-skill-list-mismatch\u002Fskills\u002Fsample\u002FSKILL.md",[],[],{"skillNo":123,"name":107,"nameZh":6,"description":124,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":125,"categories":126,"keywords":127,"publishedAt":19},"01M79PR53256JNXJ6RXXXJDTNV","\"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\"","https:\u002F\u002Fgithub.com\u002Fnick-neely\u002Fagent-skills\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fquoted-overlong-folded-description\u002Fskills\u002Fsample\u002FSKILL.md",[],[]]