[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M74JH59H7FJZW0VR6EXMHQJ5":3,"skill-same:01M74JH59H7FJZW0VR6EXMHQJ5":61},{"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":29,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":46},"01M74JH59H7FJZW0VR6EXMHQJ5","stale-deploy",null,"Use when asking \"is it live\", a deploy looks stale, or a poll loop exits instantly. Verify a GitHub Pages or CDN deploy serves new content: poll a marker new to this deploy.","MrBinnacle\u002Fskills","MrBinnacle","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F124333270?v=4","01M74JFP3ZCYZP94CVBY1JR1E6","Matthew Gruber","01M74JFP4K4A75MGTZBC9D836G",1,"MIT","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fstale-deploy\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: stale-deploy\ndescription: 'Use when asking \"is it live\", a deploy looks stale, or a poll loop exits instantly. Verify a GitHub Pages or CDN deploy serves new content: poll a marker new to this deploy.'\ndisable-model-invocation: true\n---\n\n# GitHub Pages deploy verification\n\n## Problem\n\nThree failure modes when verifying that a deploy is live via `curl`:\n\n1. **False-positive poll predicate.** Polling `until curl ... | grep -q \"\u003Cselector>\"; do sleep; done`\n   succeeds instantly if the selector existed before the deploy too. The loop exits on the OLD\n   content; you think you've verified the new state and you haven't.\n\n2. **Blocked timing chain.** `sleep 35 && curl ...` was blocked by the Claude Code Bash tool\n   (\"Blocked: sleep N followed by:\"), cancelling parallel calls (observed 2026-05-27). The\n   docs do not describe that block; they state that a command starting with `sleep` is not\n   moved to the background at its timeout (code.claude.com tools-reference, read 2026-10-17).\n\n3. **Unbounded failure.** A missing marker can mean a slow deploy, failed build, HTTP error, or\n   stale content. An endless loop distinguishes none of them and cannot return a useful verdict.\n\nThe first two bugs hit one session inside ~30 minutes. They are not in the same family but ride the\nsame task.\n\n## Context \u002F Trigger Conditions\n\nAny of:\n\n- About to push to a branch where merge = production deploy (GitHub Pages legacy mode, Netlify\n  Git, Vercel Git, etc.)\n- The platform's own deploy-status API is known to lag (e.g., `gh api repos\u002F...\u002Fpages\u002Fbuilds\u002Flatest`\n  can report stale `building` long after content is live, or report `built` before the CDN\n  catches up)\n- A task isn't \"done\" until you've confirmed the new HTML\u002FCSS\u002FJS is served, not just merged\n\n## Solution\n\n### Pick the poll predicate correctly\n\nThe string you grep for **must be content that did not exist pre-deploy**. Concretely:\n\n- **Do:** grep for a new CSS rule, new class, new token value, new copy string, new commit hash\n  echoed in a meta tag.\n- **Avoid:** grep for an element selector (`.case-footnote`, `\u003Ch1>`) that already shipped — the loop\n  exits on the cached old content.\n- **Avoid:** grep for a token name (`--accent-dim`) when only its *value* changed — the name was there\n  before too.\n\nRule: `git diff \u003Cdeployed-sha>..HEAD -- . | grep '^+'` lists candidates across every commit\nin the deploy, however many there are.\nPick one unique to the diff and likely to survive the site's build transformations.\n\n### Avoid blocked timing chains\n\nIn Claude Code's Bash tool, do not write:\n\n```bash\nsleep 35 && curl ...    # often blocked\n```\n\nUse a bounded function. Replace `deploy_status` with the platform's build-status command; it must\nprint `built` only for a successful build. The status call runs only after all attempts, so a live\ndeploy makes no extra API call.\n\n```bash\npoll_deploy() {\n  url=$1 marker=$2 max_attempts=${3:-24}; attempt=1; saw_http_error=false\n  while [ \"$attempt\" -le \"$max_attempts\" ]; do\n    body=$(curl -fsS \"$url\") && {\n      case $body in *\"$marker\"*) echo \"LIVE: new content served\"; return 0;; esac\n    } || saw_http_error=true\n    [ \"$attempt\" -eq \"$max_attempts\" ] || sleep 5\n    attempt=$((attempt + 1))\n  done\n  status=$(deploy_status 2>\u002Fdev\u002Fnull || printf unknown)\n  if $saw_http_error; then echo \"HTTP FAILURE: site returned an error (build: $status)\" >&2\n  elif [ \"$status\" = built ]; then echo \"STALE CONTENT: build succeeded but marker is absent\" >&2\n  else echo \"TIMEOUT: deploy did not complete (build: $status)\" >&2; fi\n  return 1\n}\n```\n\nThis stable contract is: bounded attempts, four named outcomes, zero only for `LIVE`, and build\nstatus deferred until the bound. The command used for `deploy_status` and the attempt count are\nadopter-specific.\n\n**Pattern A — synchronous until-loop with a check command:**\n\n```bash\npoll_deploy \"https:\u002F\u002Fyour.site\u002F\" \"\u003Cnew-content-marker>\" 24 &&\n  curl -fsS \"https:\u002F\u002Fyour.site\u002F\" | grep -E \"\u003Cverification-grep>\" | head -20\n```\n\n**Pattern B — background until-loop (preferred for slow CDNs):**\n\nInvoke `poll_deploy \"https:\u002F\u002Fyour.site\u002F\" \"\u003Cnew-content-marker>\" 60` with\n`run_in_background: true`. You get a completion notification; do not poll it yourself.\n\n### Pick the branch first\n\n- **Workflow-built Pages** (`gh api repos\u002FOWNER\u002FREPO\u002Fpages -q .build_type` prints `workflow`):\n  `pages\u002Fbuilds\u002Flatest` returns 404 (checked on this repository's site, 2026-10-17). The deploy\n  signal is the workflow run: `gh run watch \u003Crun-id> --exit-status`, then poll the marker.\n- **Already live** (the change shipped earlier): do not poll. Fetch once and compare the served\n  text to the source at the deployed commit.\n\n## Verification\n\nThe bounded poll is correct when:\n\n- `curl \u003Curl> | grep \"\u003Cmarker>\"` returns NOTHING immediately after push (proves the marker is\n  genuinely new).\n- A successful poll prints `LIVE` and returns zero; timeout, HTTP failure, and stale content each\n  print their named reason and return nonzero.\n- `deploy_status` is not called when the marker appears, and is called only after the attempt cap.\n\nIf the poll succeeds in less than ~5 seconds for a platform that typically takes 30+ seconds, your\npredicate matched pre-existing content. Re-pick the marker.\n\n## Example\n\nSession 2026-05-27, the author's public [azimuth](https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fazimuth) skill\nrepo, fixing WCAG AA contrast on `index.html`:\n\n**First attempt (wrong predicate):**\n```bash\npoll_deploy \"https:\u002F\u002Fmrbinnacle.github.io\u002Fazimuth\u002F\" \"case-footnote\" 24\n```\nLoop exited instantly. `.case-footnote` existed in the OLD HTML — only its `color` value\nchanged. False positive.\n\n**Second attempt (correct predicate):**\n```bash\npoll_deploy \"https:\u002F\u002Fmrbinnacle.github.io\u002Fazimuth\u002F\" \"overflow-x: auto\" 24\n```\n`overflow-x: auto` was a brand-new declaration on `.v-table-wrap` introduced in the same commit.\nLoop waited ~30 seconds, then exited on real new content. Verified.\n\n## Notes\n\nOpen [gotchas.md](gotchas.md) when a deploy poll exits instantly on content that pre-dated the deploy, or a `sleep && curl` chain is blocked. It records both observed failures and the anticipated ones: regional edges, rewritten markers, compressed responses, a cached 404.\n\n- The `gh api repos\u002F...\u002Fpages\u002Fbuilds\u002Flatest` endpoint is informational, not authoritative. It\n  can read `building` after the site is live, and `built` before the CDN catches up. Always\n  verify via `curl \u003Cprod-url>` instead.\n- Custom domains and CDN edges can delay propagation further (Cloudflare, Fastly; observed\n  practice, not a vendor-documented figure, 2026-10-17). Set an attempt\n  cap that covers the expected propagation window, and keep the marker deploy-unique.\n- Same pattern works for Netlify, Vercel, Cloudflare Pages, S3+CloudFront, and any other\n  static-site CDN. The harness rules apply identically.\n","# GitHub Pages deploy verification\n\n## Problem\n\nThree failure modes when verifying that a deploy is live via `curl`:\n\n1. **False-positive poll predicate.** Polling `until curl ... | grep -q \"\u003Cselector>\"; do sleep; done`\n   succeeds instantly if the selector existed before the deploy too. The loop exits on the OLD\n   content; you think you've verified the new state and you haven't.\n\n2. **Blocked timing chain.** `sleep 35 && curl ...` was blocked by the Claude Code Bash tool\n   (\"Blocked: sleep N followed by:\"), cancelling parallel calls (observed 2026-05-27). The\n   docs do not describe that block; they state that a command starting with `sleep` is not\n   moved to the background at its timeout (code.claude.com tools-reference, read 2026-10-17).\n\n3. **Unbounded failure.** A missing marker can mean a slow deploy, failed build, HTTP error, or\n   stale content. An endless loop distinguishes none of them and cannot return a useful verdict.\n\nThe first two bugs hit one session inside ~30 minutes. They are not in the same family but ride the\nsame task.\n\n## Context \u002F Trigger Conditions\n\nAny of:\n\n- About to push to a branch where merge = production deploy (GitHub Pages legacy mode, Netlify\n  Git, Vercel Git, etc.)\n- The platform's own deploy-status API is known to lag (e.g., `gh api repos\u002F...\u002Fpages\u002Fbuilds\u002Flatest`\n  can report stale `building` long after content is live, or report `built` before the CDN\n  catches up)\n- A task isn't \"done\" until you've confirmed the new HTML\u002FCSS\u002FJS is served, not just merged\n\n## Solution\n\n### Pick the poll predicate correctly\n\nThe string you grep for **must be content that did not exist pre-deploy**. Concretely:\n\n- **Do:** grep for a new CSS rule, new class, new token value, new copy string, new commit hash\n  echoed in a meta tag.\n- **Avoid:** grep for an element selector (`.case-footnote`, `\u003Ch1>`) that already shipped — the loop\n  exits on the cached old content.\n- **Avoid:** grep for a token name (`--accent-dim`) when only its *value* changed — the name was there\n  before too.\n\nRule: `git diff \u003Cdeployed-sha>..HEAD -- . | grep '^+'` lists candidates across every commit\nin the deploy, however many there are.\nPick one unique to the diff and likely to survive the site's build transformations.\n\n### Avoid blocked timing chains\n\nIn Claude Code's Bash tool, do not write:\n\n```bash\nsleep 35 && curl ...    # often blocked\n```\n\nUse a bounded function. Replace `deploy_status` with the platform's build-status command; it must\nprint `built` only for a successful build. The status call runs only after all attempts, so a live\ndeploy makes no extra API call.\n\n```bash\npoll_deploy() {\n  url=$1 marker=$2 max_attempts=${3:-24}; attempt=1; saw_http_error=false\n  while [ \"$attempt\" -le \"$max_attempts\" ]; do\n    body=$(curl -fsS \"$url\") && {\n      case $body in *\"$marker\"*) echo \"LIVE: new content served\"; return 0;; esac\n    } || saw_http_error=true\n    [ \"$attempt\" -eq \"$max_attempts\" ] || sleep 5\n    attempt=$((attempt + 1))\n  done\n  status=$(deploy_status 2>\u002Fdev\u002Fnull || printf unknown)\n  if $saw_http_error; then echo \"HTTP FAILURE: site returned an error (build: $status)\" >&2\n  elif [ \"$status\" = built ]; then echo \"STALE CONTENT: build succeeded but marker is absent\" >&2\n  else echo \"TIMEOUT: deploy did not complete (build: $status)\" >&2; fi\n  return 1\n}\n```\n\nThis stable contract is: bounded attempts, four named outcomes, zero only for `LIVE`, and build\nstatus deferred until the bound. The command used for `deploy_status` and the attempt count are\nadopter-specific.\n\n**Pattern A — synchronous until-loop with a check command:**\n\n```bash\npoll_deploy \"https:\u002F\u002Fyour.site\u002F\" \"\u003Cnew-content-marker>\" 24 &&\n  curl -fsS \"https:\u002F\u002Fyour.site\u002F\" | grep -E \"\u003Cverification-grep>\" | head -20\n```\n\n**Pattern B — background until-loop (preferred for slow CDNs):**\n\nInvoke `poll_deploy \"https:\u002F\u002Fyour.site\u002F\" \"\u003Cnew-content-marker>\" 60` with\n`run_in_background: true`. You get a completion notification; do not poll it yourself.\n\n### Pick the branch first\n\n- **Workflow-built Pages** (`gh api repos\u002FOWNER\u002FREPO\u002Fpages -q .build_type` prints `workflow`):\n  `pages\u002Fbuilds\u002Flatest` returns 404 (checked on this repository's site, 2026-10-17). The deploy\n  signal is the workflow run: `gh run watch \u003Crun-id> --exit-status`, then poll the marker.\n- **Already live** (the change shipped earlier): do not poll. Fetch once and compare the served\n  text to the source at the deployed commit.\n\n## Verification\n\nThe bounded poll is correct when:\n\n- `curl \u003Curl> | grep \"\u003Cmarker>\"` returns NOTHING immediately after push (proves the marker is\n  genuinely new).\n- A successful poll prints `LIVE` and returns zero; timeout, HTTP failure, and stale content each\n  print their named reason and return nonzero.\n- `deploy_status` is not called when the marker appears, and is called only after the attempt cap.\n\nIf the poll succeeds in less than ~5 seconds for a platform that typically takes 30+ seconds, your\npredicate matched pre-existing content. Re-pick the marker.\n\n## Example\n\nSession 2026-05-27, the author's public [azimuth](https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fazimuth) skill\nrepo, fixing WCAG AA contrast on `index.html`:\n\n**First attempt (wrong predicate):**\n```bash\npoll_deploy \"https:\u002F\u002Fmrbinnacle.github.io\u002Fazimuth\u002F\" \"case-footnote\" 24\n```\nLoop exited instantly. `.case-footnote` existed in the OLD HTML — only its `color` value\nchanged. False positive.\n\n**Second attempt (correct predicate):**\n```bash\npoll_deploy \"https:\u002F\u002Fmrbinnacle.github.io\u002Fazimuth\u002F\" \"overflow-x: auto\" 24\n```\n`overflow-x: auto` was a brand-new declaration on `.v-table-wrap` introduced in the same commit.\nLoop waited ~30 seconds, then exited on real new content. Verified.\n\n## Notes\n\nOpen [gotchas.md](gotchas.md) when a deploy poll exits instantly on content that pre-dated the deploy, or a `sleep && curl` chain is blocked. It records both observed failures and the anticipated ones: regional edges, rewritten markers, compressed responses, a cached 404.\n\n- The `gh api repos\u002F...\u002Fpages\u002Fbuilds\u002Flatest` endpoint is informational, not authoritative. It\n  can read `building` after the site is live, and `built` before the CDN catches up. Always\n  verify via `curl \u003Cprod-url>` instead.\n- Custom domains and CDN edges can delay propagation further (Cloudflare, Fastly; observed\n  practice, not a vendor-documented figure, 2026-10-17). Set an attempt\n  cap that covers the expected propagation window, and keep the marker deploy-unique.\n- Same pattern works for Netlify, Vercel, Cloudflare Pages, S3+CloudFront, and any other\n  static-site CDN. The harness rules apply identically.\n",{"name":5,"description":7,"disable-model-invocation":24},true,"main","fb8ab02da6c849e2cdb704371f5c8c190512d15e","skills\u002Fengineering\u002Fstale-deploy","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills","https:\u002F\u002Fmrbinnacle.github.io\u002Fskills\u002F","https:\u002F\u002Fgithub.com\u002FMrBinnacle","https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmatthew-g-83390523","I make AI coding agents work in messy, real environments, and measure whether they did. Ex-Army NCO, ex-J.P. Morgan.","2026-11-15 04:37:29",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":24},[36,39,42],{"tool":37,"command":38},"npx","npx skills add MrBinnacle\u002Fskills --skill stale-deploy",{"tool":40,"command":41},"bunx","bunx skills add MrBinnacle\u002Fskills --skill stale-deploy",{"tool":43,"command":44},"pnpm","pnpm dlx skills add MrBinnacle\u002Fskills --skill stale-deploy","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fstale-deploy\u002FSKILL.md",[47,53,57],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"evals\u002Fevals.json","REFERENCE","参考文档",2501,"https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fstale-deploy\u002Fevals\u002Fevals.json",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"EVIDENCE.md",4826,"https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fstale-deploy\u002FEVIDENCE.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"gotchas.md",3206,"https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fstale-deploy\u002Fgotchas.md",[62,69,76,83,90,97],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M74JHKQKA5V9N4B4VDAHBGAP","subagent-handback","Use when prompting, briefing or dispatching a research subagent, or when one goes idle with no report. Name a return channel, check its tool grant, verify every claim, negatives first.","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Forchestration\u002Fsubagent-handback\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M74JHGXP4YFAGQCRTMFQ8Q5F","disposition-schema","Use when dispatching parallel agents to verify or adjudicate findings (adversarial verify, security disposition, council fires), or before you consolidate their outputs and risk an ID collision.","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Forchestration\u002Fdisposition-schema\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M74JHDMFE7VN27QTC6YRPW22","decision-rights","Use when writing a handoff, plan, ADR or subagent brief, or one says \"do not re-litigate\" or \"final\". Separates evidence access from decision rights and marks what is non-negotiable.","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Forchestration\u002Fdecision-rights\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M74JHARZJ246DYV6SSXFRNXT","dead-predicate","Use when a hook didn't fire, a reminder never arrived, or before claiming a rule is hook-enforced: a live router rule can match nothing (missing word, JSON escape, shadowing, stale name).","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fmeta\u002Fdead-predicate\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M74JH7P66AGN3RHKH7DVA5Y4","vacuous-check","Fail-open success checks. Use when a script says OK but nothing happened, a probe batch returns a false negative or 0 results, or `String(got) === String(want)` passes on a type mismatch.","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fvacuous-check\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M74JH297TR1CZVQHRPBEXGFT","pull-rebase","Use before `git pull` where `pull.rebase` may be `true`, or after a pull rebased your commits. `--no-ff` does not stop the rebase; every local SHA changes. Check config first; recover by reflog.","https:\u002F\u002Fgithub.com\u002FMrBinnacle\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fpull-rebase\u002FSKILL.md",[],[]]