[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AC9ZPAHTCAXYP2K6HQFP4P":3,"skill-same:01M7AC9ZPAHTCAXYP2K6HQFP4P":48},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":14,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":47},"01M7AC9ZPAHTCAXYP2K6HQFP4P","speckit-verify-review-ship-verify",null,"Run post-convergence executable gates and diff hygiene checks","cadugevaerd\u002Fgrill-with-docs","cadugevaerd","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F63358082?v=4","01M7AC7XM4CZJ31SYHN7K3WARC","Carlos Eduardo Gevaerd Araujo","01M7AC7XMXXNZBHQNVQKVN8DB7",0,"MIT","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fspeckit-verify-review-ship-verify\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: speckit-verify-review-ship-verify\ndescription: Run post-convergence executable gates and diff hygiene checks\ncompatibility: Requires spec-kit project structure with .specify\u002F directory\nmetadata:\n  author: github-spec-kit\n  source: verify-review-ship:commands\u002Fverify.md\n---\n\n# Verify Gate\n\nRun only after the official `\u002Fspeckit.converge` reports `Converged`. This command proves the current implementation is executable; it does not decide whether implementation fulfills the feature intent.\n\n## User Input\n\n$ARGUMENTS\n\n## Converge Handoff (Required)\n\nRequire evidence from the current host session\u002Fhistory that official `\u002Fspeckit.converge` ran after the latest `\u002Fspeckit.implement`.\n\n```text\nOutcome: converged | tasks_appended | unavailable\nSource fingerprint: reviewed-scope tree hash + uncommitted diff hash + tasks.md hash\nEvidence source: official command result in the current session\u002Fhistory\n```\n\n- `tasks_appended`: return `VERIFY: BLOCKED`; run `\u002Fspeckit.implement`, then `\u002Fspeckit.converge` again.\n- `unavailable`, missing, ambiguous, or stale evidence: return `VERIFY: BLOCKED`; run `\u002Fspeckit.converge` again.\n- Capture the current source fingerprint in this report, computed as defined in **Source Fingerprint (Canonical)** below. A changed fingerprint makes the report stale; a report written or committed by these gates MUST NOT change it.\n\nThe official Converge command owns spec\u002Fplan\u002Ftasks-to-code completeness. This command **must not reconstruct the intent inventory** and **must not repeat spec-to-code completeness analysis**. Do not enumerate FRs, acceptance scenarios, plan decisions, or tasks as an independent pass\u002Ffail ledger.\n\n## Source Fingerprint (Canonical)\n\nThe fingerprint identifies the **reviewed content**, never the commit that recorded a gate\nreport. It is computed by `.specify\u002Fextensions\u002Fverify-review-ship\u002Fscripts\u002Fsource-fingerprint.sh`, which is the single implementation —\ndo not restate the algorithm in prose:\n\n```bash\n.specify\u002Fextensions\u002Fverify-review-ship\u002Fscripts\u002Fsource-fingerprint.sh \u003Cfeature-dir> [config.yml]\n```\n\nIt emits the three components and the combined value:\n\n```text\ntree \u003Csha256>          committed and staged content of the reviewed scope\nwork \u003Csha256>          uncommitted changes AND untracked files in that scope\nplan \u003Csha256>          the feature's tasks.md\nfingerprint \u003Ctree>-\u003Cwork>-\u003Cplan>\n```\n\nReport the three components, not only the combined value, so a mismatch says which part moved.\n\n**Exclusions come from `converge.fingerprint_exclude`** in the extension configuration, so the\npublished setting and the algorithm share one effective set. The script falls back to its\ndocumented defaults only when the key is absent. A project that stores gate reports elsewhere\nchanges the configuration, not the script; a report path inside the fingerprint reintroduces the\ndefect below.\n\n**Why `HEAD` is not part of it.** Verify and Review are told to write their reports into the\nrepository. Committing a report changes `HEAD` without changing anything that was reviewed, so a\n`HEAD`-based fingerprint invalidates itself between two gates of the same run and forces\n`BLOCKED` on work that never moved.\n\n**Why untracked files are part of it.** `git diff HEAD` does not see a file that was never\nadded. An implementation delivered as a new, unstaged file would otherwise be invisible to the\nfingerprint, letting unreviewed content pass a matching-evidence check.\n\n## Scope Discovery\n\nResolve repository root, active feature, extension configuration, current diff, CI\u002Ftask-runner\u002Fmanifests, quickstart, executable contracts, and available project gates. Missing executable context or a forbidden gate is `BLOCKED`; an absent optional gate is `SKIPPED` with evidence.\n\n## Execution\n\nDiscover workers\u002Fsubagents when supported; otherwise use sequential fallback. Assign at most one read-only validator to each independent executable gate or hygiene item. Validators never edit, stage, commit, deploy, or decide the verdict.\n\nRun safe canonical gates in this order when applicable: tests, build, lint, typecheck, format, security, quickstart\u002Fcontracts, and required manual gates. Record exact command, result, evidence, and validator.\n\nInspect diff hygiene: changed\u002Fgenerated\u002Funrelated files, accidental secrets or environment files, and whether executable scenarios have supporting tests. Never stage or commit.\n\n## Report\n\nWrite when allowed to `.specify\u002Freports\u002Fverify-review-ship\u002Fverify.md` and the active feature directory.\n\n```markdown\n## Verify Report\n\nVerdict: PASS | FAIL | BLOCKED\nSource fingerprint: tree \u003Csha> \u002F work \u003Csha> \u002F plan \u003Csha>   (gate reports excluded)\nConverge: CONVERGED | STALE | MISSING | TASKS_APPENDED\n\n### Operational Gates\n| Gate | Command | Result | Evidence | Validator |\n\n### Diff Hygiene\n### Executable Scenarios\n### Failures \u002F Blockers\n\n### Next Action\n- PASS: run `\u002Fspeckit.verify-review-ship.review`\n- FAIL\u002FBLOCKED: fix the listed operational issue; if code or Spec Kit artifacts changed, rerun `\u002Fspeckit.converge` first.\n```\n\n## Decision Rules\n\n`PASS` requires fresh `CONVERGED` evidence and every mandatory executable gate passing. Never infer a gate pass from another result. A source change invalidates the report.","# Verify Gate\n\nRun only after the official `\u002Fspeckit.converge` reports `Converged`. This command proves the current implementation is executable; it does not decide whether implementation fulfills the feature intent.\n\n## User Input\n\n$ARGUMENTS\n\n## Converge Handoff (Required)\n\nRequire evidence from the current host session\u002Fhistory that official `\u002Fspeckit.converge` ran after the latest `\u002Fspeckit.implement`.\n\n```text\nOutcome: converged | tasks_appended | unavailable\nSource fingerprint: reviewed-scope tree hash + uncommitted diff hash + tasks.md hash\nEvidence source: official command result in the current session\u002Fhistory\n```\n\n- `tasks_appended`: return `VERIFY: BLOCKED`; run `\u002Fspeckit.implement`, then `\u002Fspeckit.converge` again.\n- `unavailable`, missing, ambiguous, or stale evidence: return `VERIFY: BLOCKED`; run `\u002Fspeckit.converge` again.\n- Capture the current source fingerprint in this report, computed as defined in **Source Fingerprint (Canonical)** below. A changed fingerprint makes the report stale; a report written or committed by these gates MUST NOT change it.\n\nThe official Converge command owns spec\u002Fplan\u002Ftasks-to-code completeness. This command **must not reconstruct the intent inventory** and **must not repeat spec-to-code completeness analysis**. Do not enumerate FRs, acceptance scenarios, plan decisions, or tasks as an independent pass\u002Ffail ledger.\n\n## Source Fingerprint (Canonical)\n\nThe fingerprint identifies the **reviewed content**, never the commit that recorded a gate\nreport. It is computed by `.specify\u002Fextensions\u002Fverify-review-ship\u002Fscripts\u002Fsource-fingerprint.sh`, which is the single implementation —\ndo not restate the algorithm in prose:\n\n```bash\n.specify\u002Fextensions\u002Fverify-review-ship\u002Fscripts\u002Fsource-fingerprint.sh \u003Cfeature-dir> [config.yml]\n```\n\nIt emits the three components and the combined value:\n\n```text\ntree \u003Csha256>          committed and staged content of the reviewed scope\nwork \u003Csha256>          uncommitted changes AND untracked files in that scope\nplan \u003Csha256>          the feature's tasks.md\nfingerprint \u003Ctree>-\u003Cwork>-\u003Cplan>\n```\n\nReport the three components, not only the combined value, so a mismatch says which part moved.\n\n**Exclusions come from `converge.fingerprint_exclude`** in the extension configuration, so the\npublished setting and the algorithm share one effective set. The script falls back to its\ndocumented defaults only when the key is absent. A project that stores gate reports elsewhere\nchanges the configuration, not the script; a report path inside the fingerprint reintroduces the\ndefect below.\n\n**Why `HEAD` is not part of it.** Verify and Review are told to write their reports into the\nrepository. Committing a report changes `HEAD` without changing anything that was reviewed, so a\n`HEAD`-based fingerprint invalidates itself between two gates of the same run and forces\n`BLOCKED` on work that never moved.\n\n**Why untracked files are part of it.** `git diff HEAD` does not see a file that was never\nadded. An implementation delivered as a new, unstaged file would otherwise be invisible to the\nfingerprint, letting unreviewed content pass a matching-evidence check.\n\n## Scope Discovery\n\nResolve repository root, active feature, extension configuration, current diff, CI\u002Ftask-runner\u002Fmanifests, quickstart, executable contracts, and available project gates. Missing executable context or a forbidden gate is `BLOCKED`; an absent optional gate is `SKIPPED` with evidence.\n\n## Execution\n\nDiscover workers\u002Fsubagents when supported; otherwise use sequential fallback. Assign at most one read-only validator to each independent executable gate or hygiene item. Validators never edit, stage, commit, deploy, or decide the verdict.\n\nRun safe canonical gates in this order when applicable: tests, build, lint, typecheck, format, security, quickstart\u002Fcontracts, and required manual gates. Record exact command, result, evidence, and validator.\n\nInspect diff hygiene: changed\u002Fgenerated\u002Funrelated files, accidental secrets or environment files, and whether executable scenarios have supporting tests. Never stage or commit.\n\n## Report\n\nWrite when allowed to `.specify\u002Freports\u002Fverify-review-ship\u002Fverify.md` and the active feature directory.\n\n```markdown\n## Verify Report\n\nVerdict: PASS | FAIL | BLOCKED\nSource fingerprint: tree \u003Csha> \u002F work \u003Csha> \u002F plan \u003Csha>   (gate reports excluded)\nConverge: CONVERGED | STALE | MISSING | TASKS_APPENDED\n\n### Operational Gates\n| Gate | Command | Result | Evidence | Validator |\n\n### Diff Hygiene\n### Executable Scenarios\n### Failures \u002F Blockers\n\n### Next Action\n- PASS: run `\u002Fspeckit.verify-review-ship.review`\n- FAIL\u002FBLOCKED: fix the listed operational issue; if code or Spec Kit artifacts changed, rerun `\u002Fspeckit.converge` first.\n```\n\n## Decision Rules\n\n`PASS` requires fresh `CONVERGED` evidence and every mandatory executable gate passing. Never infer a gate pass from another result. A source change invalidates the report.",{"name":5,"metadata":24,"description":7,"compatibility":27},{"author":25,"source":26},"github-spec-kit","verify-review-ship:commands\u002Fverify.md","Requires spec-kit project structure with .specify\u002F directory","main","dd561c2e23614861198af1b46cc720d1b0fb2cf0",".agents\u002Fskills\u002Fspeckit-verify-review-ship-verify","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs","https:\u002F\u002Fgithub.com\u002Fcadugevaerd","2026-09-28 01:22:45",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add cadugevaerd\u002Fgrill-with-docs --skill speckit-verify-review-ship-verify",{"tool":40,"command":41},"bunx","bunx skills add cadugevaerd\u002Fgrill-with-docs --skill speckit-verify-review-ship-verify",{"tool":43,"command":44},"pnpm","pnpm dlx skills add cadugevaerd\u002Fgrill-with-docs --skill speckit-verify-review-ship-verify","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fspeckit-verify-review-ship-verify\u002FSKILL.md",true,[],[49,55,62,69,76,81],{"skillNo":50,"name":51,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7ACCWNYYQVH259P3X96A3PS","offline-presentation-fixture","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Forchestration\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M7ACCS4KGQ0J2V3T60DTETGD","grill-with-docs","Entrevista decisões arquiteturais por work item isolado, mantém feature plan-only e oferece hotfix-fast executável com HOTFIX-GO fail-closed.","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fgrill-with-docs\u002FSKILL.md",[],[],{"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},"01M7ACCMCXGYXGXSM9FXNECN2G","grill-partition","Particiona tasks.md em subfases file-disjuntas e emite o Execution DAG versionado da etapa partition, de forma determinística e sem julgamento de modelo.","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fgrill-partition\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},"01M7ACCH73N28NESB337VHEC0R","grill-implement-parallel","Orquestra os workers não-frontier da etapa implement-parallel sobre o Execution DAG, com worktree isolado por worker e receipt submetido apenas pelo leader.","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002Fplugin\u002Fskills\u002Fgrill-implement-parallel\u002FSKILL.md",[],[],{"skillNo":77,"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":78,"categories":79,"keywords":80,"publishedAt":19},"01M7ACCDSD9CGXTJDA87C269NZ","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fspeckit-verify-review-ship-verify\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"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":84,"categories":85,"keywords":86,"publishedAt":19},"01M7ACCBFXR7JX9RN0PGRAZTA0","speckit-verify-review-ship-v","https:\u002F\u002Fgithub.com\u002Fcadugevaerd\u002Fgrill-with-docs\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fspeckit-verify-review-ship-v\u002FSKILL.md",[],[]]