[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77WCFKXF5Y3JKX6QN37KVTF":3,"skill-same:01M77WCFKXF5Y3JKX6QN37KVTF":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":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M77WCFKXF5Y3JKX6QN37KVTF","autoreview",null,"Closeout code review with a structured second-model pass. Use after non-trivial changes or when the user asks for autoreview, Codex review, or a review panel.","vana-com\u002Fvana-cli","vana-com","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F123112561?v=4","01M77WCAG7S42PDY7VTEVWXQD1","Vana","01M77WCAGV436KWGZWHNHZXGVH",2,"MIT","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fautoreview\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: autoreview\ndescription: \"Closeout code review with a structured second-model pass. Use after non-trivial changes or when the user asks for autoreview, Codex review, or a review panel.\"\n---\n\n# Auto Review\n\nRun the repo's bundled structured review helper after the nearest owning proof.\nCodex is the default reviewer; set `AUTOREVIEW_ENGINE=claude` (or pass\n`--engine claude`) when Codex is unavailable. Review findings are advisory; the\ntask, owning code and contracts, and their nearest checks remain authoritative.\n\n## Closeout\n\n1. Record the baseline: user request, violated invariant, owner boundary,\n   intended behavior, and target ref. Completion means a finding can be\n   classified without silently broadening the task.\n2. Run the nearest focused checks with the repo's own scripts. Completion means\n   the pre-review proof is current. Full proof for this repo:\n\n   ```bash\n   pnpm validate\n   ```\n\n   - This repo is pnpm-only; never suggest `npm` commands here. `pnpm validate` is `pnpm lint` (`tsc --noEmit`) + `pnpm lint:eslint` + `pnpm format:check` + `pnpm test`, which is exactly what the CI `test` job runs.\n   - Focused tests: `pnpm test test\u002F\u003Carea>\u002F\u003Cfile>.test.ts`. E2E lives behind `pnpm test:e2e`.\n   - CLI surface changes also carry `pnpm preflight:cli` (focused tests, build, format check, regenerated demo transcripts).\n   - Packaging and install lanes are proved by `pnpm build && pnpm pack:check`, `pnpm build:sea:smoke`, `pnpm sea:check`, and `pnpm test:install:unix`; run them only when the change touches build, packaging, or install.\n   - `connect\u002F` and `examples\u002F*` are separate workspace packages that root CI path-ignores. Prove them from inside the package.\n   - The commit hook runs commitlint and lint-staged, so a commit message must be a conventional commit.\n\n3. Select the review target:\n\n   ```bash\n   # Dirty staged, unstaged, and untracked work\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode local\n\n   # Feature\u002FPR branch; PRs in this repo target main\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode branch --base origin\u002Fmain\n\n   # One committed change\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode commit --commit HEAD\n\n   # Exact commit-to-commit diff (no merge-base expansion)\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode range --base \u003Cbase> --head \u003Chead>\n   ```\n\n   For an open PR, use its actual base (the helper asks `gh pr view` when no\n   `--base` is given). Branch mode requires a clean worktree; local changes\n   belong in local mode. Commit mode reviews merge commits against their first\n   parent. A clean local review proves only that no local patch exists.\n\n4. Adjudicate every finding against the real code path, adjacent files,\n   dependency contracts, and repo doctrine. Completion means each finding is\n   accepted with evidence or rejected with a one-line reason.\n5. Classify accepted findings before editing:\n   - **In-scope blocker**: same invariant and owner neighborhood; fix it.\n   - **Follow-up**: real, but a different owner or bug class; report it.\n   - **Stop-and-escalate**: changes protocol, on-disk state, exported API, CLI\n     flags or exit codes, packaging, release process, or another owner's\n     boundary; request direction.\n\n6. After an in-scope fix, rerun affected focused proof and autoreview. After two\n   review-triggered patch cycles without convergence, pause and reclassify all\n   remaining findings. Continue only when every remaining item is still an\n   in-scope blocker.\n7. Stop on the first clean helper exit. Report the command, proof run,\n   accepted\u002Frejected findings, and clean result.\n\n## Finding Threshold\n\nThe helper defaults to P0: issues that block the current change because they\nmaterially break its normal flow, outcome, or safety boundary. Broaden only on\nexplicit request:\n\n```bash\n.agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode branch --base origin\u002Fmain --max-priority P1\n```\n\nP2\u002FP3 polish and speculative hardening are not blockers by default. Security\nfindings must name a concrete risk or removed safety check; security-adjacent\nfunctionality alone is not a defect.\n\n## Repo Boundaries\n\n- The touched path and owning code select scope; the nearest real oracle proves\n  behavior. A clean source review never replaces required runtime, integration,\n  e2e, or generated-file proof.\n- Fix the root cause across relevant siblings inside the same owner boundary.\n  Keep unrelated cleanup out of the landing lane.\n- PRs target `main`; semantic-release publishes the npm package and the release\n  assets from `main`, and the `semantic-pull-request` check requires a\n  conventional-commit PR title. Review never authorizes push, merge, release, or\n  tag actions.\n- On release, hotfix, or publish work, apply freeze discipline: fix only the\n  release blocker, exact backport, crash, data-loss, install\u002Fupgrade breakage,\n  or concrete security exposure. Route other findings to follow-up work.\n- The published surface is the CLI binary plus `@opendatalabs\u002Fconnect` entry\n  points. Treat exported types, CLI flags, exit codes, and install scripts as\n  public contract: a finding that changes them is stop-and-escalate, not an\n  in-scope fix.\n- Prose-only skill\u002Fdocs changes may skip the model review, but still run their\n  nearest lightweight check (Prettier where it covers Markdown). Executable\n  examples, scripts, configuration, generated files, and user-facing docs do\n  not qualify.\n\n## Panels\n\nPanels are opt-in. Use them only when requested or when the risk justifies the\nextra cost:\n\n```bash\n.agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode branch --base origin\u002Fmain --panel\n```\n\nThe main agent still adjudicates every merged finding. Never push merely to make\na review possible.\n\n## Self-check\n\nThe helper's target-selection logic has unit tests; run them after editing the\nscript:\n\n```bash\npython3 .agents\u002Fskills\u002Fautoreview\u002Ftests\u002Ftest_autoreview.py\n```\n\n`scripts\u002Ftest-review-harness` builds a throwaway repo with a deliberately unsafe\npatch (or a safe security-adjacent one) and checks that a selected engine\nreports, or correctly does not report, findings.\n","# Auto Review\n\nRun the repo's bundled structured review helper after the nearest owning proof.\nCodex is the default reviewer; set `AUTOREVIEW_ENGINE=claude` (or pass\n`--engine claude`) when Codex is unavailable. Review findings are advisory; the\ntask, owning code and contracts, and their nearest checks remain authoritative.\n\n## Closeout\n\n1. Record the baseline: user request, violated invariant, owner boundary,\n   intended behavior, and target ref. Completion means a finding can be\n   classified without silently broadening the task.\n2. Run the nearest focused checks with the repo's own scripts. Completion means\n   the pre-review proof is current. Full proof for this repo:\n\n   ```bash\n   pnpm validate\n   ```\n\n   - This repo is pnpm-only; never suggest `npm` commands here. `pnpm validate` is `pnpm lint` (`tsc --noEmit`) + `pnpm lint:eslint` + `pnpm format:check` + `pnpm test`, which is exactly what the CI `test` job runs.\n   - Focused tests: `pnpm test test\u002F\u003Carea>\u002F\u003Cfile>.test.ts`. E2E lives behind `pnpm test:e2e`.\n   - CLI surface changes also carry `pnpm preflight:cli` (focused tests, build, format check, regenerated demo transcripts).\n   - Packaging and install lanes are proved by `pnpm build && pnpm pack:check`, `pnpm build:sea:smoke`, `pnpm sea:check`, and `pnpm test:install:unix`; run them only when the change touches build, packaging, or install.\n   - `connect\u002F` and `examples\u002F*` are separate workspace packages that root CI path-ignores. Prove them from inside the package.\n   - The commit hook runs commitlint and lint-staged, so a commit message must be a conventional commit.\n\n3. Select the review target:\n\n   ```bash\n   # Dirty staged, unstaged, and untracked work\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode local\n\n   # Feature\u002FPR branch; PRs in this repo target main\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode branch --base origin\u002Fmain\n\n   # One committed change\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode commit --commit HEAD\n\n   # Exact commit-to-commit diff (no merge-base expansion)\n   .agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode range --base \u003Cbase> --head \u003Chead>\n   ```\n\n   For an open PR, use its actual base (the helper asks `gh pr view` when no\n   `--base` is given). Branch mode requires a clean worktree; local changes\n   belong in local mode. Commit mode reviews merge commits against their first\n   parent. A clean local review proves only that no local patch exists.\n\n4. Adjudicate every finding against the real code path, adjacent files,\n   dependency contracts, and repo doctrine. Completion means each finding is\n   accepted with evidence or rejected with a one-line reason.\n5. Classify accepted findings before editing:\n   - **In-scope blocker**: same invariant and owner neighborhood; fix it.\n   - **Follow-up**: real, but a different owner or bug class; report it.\n   - **Stop-and-escalate**: changes protocol, on-disk state, exported API, CLI\n     flags or exit codes, packaging, release process, or another owner's\n     boundary; request direction.\n\n6. After an in-scope fix, rerun affected focused proof and autoreview. After two\n   review-triggered patch cycles without convergence, pause and reclassify all\n   remaining findings. Continue only when every remaining item is still an\n   in-scope blocker.\n7. Stop on the first clean helper exit. Report the command, proof run,\n   accepted\u002Frejected findings, and clean result.\n\n## Finding Threshold\n\nThe helper defaults to P0: issues that block the current change because they\nmaterially break its normal flow, outcome, or safety boundary. Broaden only on\nexplicit request:\n\n```bash\n.agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode branch --base origin\u002Fmain --max-priority P1\n```\n\nP2\u002FP3 polish and speculative hardening are not blockers by default. Security\nfindings must name a concrete risk or removed safety check; security-adjacent\nfunctionality alone is not a defect.\n\n## Repo Boundaries\n\n- The touched path and owning code select scope; the nearest real oracle proves\n  behavior. A clean source review never replaces required runtime, integration,\n  e2e, or generated-file proof.\n- Fix the root cause across relevant siblings inside the same owner boundary.\n  Keep unrelated cleanup out of the landing lane.\n- PRs target `main`; semantic-release publishes the npm package and the release\n  assets from `main`, and the `semantic-pull-request` check requires a\n  conventional-commit PR title. Review never authorizes push, merge, release, or\n  tag actions.\n- On release, hotfix, or publish work, apply freeze discipline: fix only the\n  release blocker, exact backport, crash, data-loss, install\u002Fupgrade breakage,\n  or concrete security exposure. Route other findings to follow-up work.\n- The published surface is the CLI binary plus `@opendatalabs\u002Fconnect` entry\n  points. Treat exported types, CLI flags, exit codes, and install scripts as\n  public contract: a finding that changes them is stop-and-escalate, not an\n  in-scope fix.\n- Prose-only skill\u002Fdocs changes may skip the model review, but still run their\n  nearest lightweight check (Prettier where it covers Markdown). Executable\n  examples, scripts, configuration, generated files, and user-facing docs do\n  not qualify.\n\n## Panels\n\nPanels are opt-in. Use them only when requested or when the risk justifies the\nextra cost:\n\n```bash\n.agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview --mode branch --base origin\u002Fmain --panel\n```\n\nThe main agent still adjudicates every merged finding. Never push merely to make\na review possible.\n\n## Self-check\n\nThe helper's target-selection logic has unit tests; run them after editing the\nscript:\n\n```bash\npython3 .agents\u002Fskills\u002Fautoreview\u002Ftests\u002Ftest_autoreview.py\n```\n\n`scripts\u002Ftest-review-harness` builds a throwaway repo with a deliberately unsafe\npatch (or a safe security-adjacent one) and checks that a selected engine\nreports, or correctly does not report, findings.\n",{"name":5,"description":7},"main","a55d0a0006da6a743b6452796435afa6ed177dd8",".agents\u002Fskills\u002Fautoreview",0,"https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli","https:\u002F\u002Fgithub.com\u002Fvana-com","https:\u002F\u002Fvana.org","The first open protocol for data sovereignty.","2026-11-13 18:19:18",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add vana-com\u002Fvana-cli --skill autoreview",{"tool":39,"command":40},"bunx","bunx skills add vana-com\u002Fvana-cli --skill autoreview",{"tool":42,"command":43},"pnpm","pnpm dlx skills add vana-com\u002Fvana-cli --skill autoreview","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fautoreview\u002FSKILL.md",true,[47,53,57],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"scripts\u002Fautoreview","SCRIPT","脚本",40914,"https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Fautoreview",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"scripts\u002Ftest-review-harness",4498,"https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fautoreview\u002Fscripts\u002Ftest-review-harness",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"tests\u002Ftest_autoreview.py",3893,"https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fautoreview\u002Ftests\u002Ftest_autoreview.py",[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},"01M7BD5YJ16KD0X3M379XA2A4A","verify-setup-resume","Verify BUI-936 setup\u002Fresume identities, funding readiness and overlapping resumes. Use when changing app request, test-credit transport or the request journal. Runs bounded TLA+ checks and offline real-code replays without live funding or paid reads.","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fverify-setup-resume\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},"01M7BD5V2155AM7W1PWRBY237Y","verify-server-supervision","Verify the local Personal Server stop and restart boundary. Use when changing the supervisor's backoff, port probing, replacement launch or Stop handling.","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fverify-server-supervision\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},"01M7BD5PPVKFVVGS9B846B99ES","verify-paid-read-recovery","Verify interrupted and overlapping CLI paid reads, receipt-file writers and dead-owner recovery. Use when changing src\u002Fcli\u002Fapp\u002Fread.ts, src\u002Fcore\u002Freceipts.ts or src\u002Fcore\u002Freceipts-lock.ts. Runs TLA+ models and offline real-code replays without live paid reads.","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fverify-paid-read-recovery\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},"01M7BD5JDEGV165NFZJWSWJBN4","verify-mainnet-registration","Verify mainnet app registration and Account association recovery. Use for the registration journal, gateway confirmation, retry, or ownership check.","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fverify-mainnet-registration\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},"01M7BD5ETK33JAD1F1XQXGPYYG","verify-hosted-provisioning","Verify hosted Personal Server resource ownership across create and delete. Use for Connect's server API, provisioning-row updates, or deprovisioning admission.","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fverify-hosted-provisioning\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},"01M7BD5BBKFHJ86JFBE8KT1S78","verify-export-settlement","Verify queued export owner binding and generation settlement. Use for result publication, pending upload receipts, or retry and sync changes.","https:\u002F\u002Fgithub.com\u002Fvana-com\u002Fvana-cli\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fverify-export-settlement\u002FSKILL.md",[],[]]