[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EVPD6J9ETPCSW3GWH5YXAH":3,"skill-same:01M7EVPD6J9ETPCSW3GWH5YXAH":45},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":13,"repositoryHtmlUrl":26,"repositoryHomepageUrl":6,"ownerHtmlUrl":27,"ownerBlogUrl":28,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7EVPD6J9ETPCSW3GWH5YXAH","new-lmfdb-type-policy",null,"tdupu\u002Fagent-skills-public","tdupu","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F3440403?v=4","01M7EV32X9J0EFE5T4P8TKQTCQ","Taylor Dupuy","01M7EV32XZ1Z4MDBYAFRCJD3BP",0,"GPL-3.0","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fskills\u002Fnew-lmfdb-type-policy\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: new-lmfdb-type-policy\ndescription: Create a new LMFDB type the policy-conformant way — gate the justification (LM4.1), design and freeze the label scheme first (LM1.x, LM4.4), then drive create-lmfdb-type + update-schema, record the webpage decision, and finish with a check-lmfdb-hygiene audit. Use when the user says \"new lmfdb type\", \"add an LMFDB type for X the right way\", \"policy-gated type creation\", \"new-lmfdb-type-policy\", or whenever create-lmfdb-type is about to run on a type that doesn't have an approved label scheme and justification yet. Companion to check-lmfdb-hygiene; wraps (does not replace) create-lmfdb-type. NOTE: despite the -policy suffix, this skill creates LMFDB TYPES following the policy, not policy rules — amendments to lmfdb\u002FPOLICY.md go through a human-gated proposal per POLICY-POLICY.md PP1.4 (new-hygiene-policy format), same caveat as new-math-bead-policy.\n---\n\n# new-lmfdb-type-policy\n\nPolicy-gated wrapper around [[create-lmfdb-type]]. That skill scaffolds\nthe code; THIS skill makes sure the type deserves to exist, its label\nscheme is right before any code is written, and the result passes\n[POLICY.md](..\u002F..\u002FPOLICY.md) before the bead closes.\n\n**Why the wrapper exists:** labels are permanent (LM1.7) and a type is a\nseven-file contract (LM4.3). The expensive mistakes — wrong label\nscheme, type-that-should-have-been-columns, unrecorded webpage decision\n— all happen *before* `create-lmfdb-type`'s Step 1. This skill front-\nloads those decisions and gates them on the human adjudicator.\n\n## Step 0 — Read the policy and the dependency check\n\n```bash\ncat \u003Cmathcity-pack-root>\u002Fsubdomains\u002Flmfdb\u002FPOLICY.md\n```\n\nThen verify the pipeline conf exists before proceeding (P1.14):\n\n```bash\n# Discover conf: project root first, hecke fallback\nCONF=\"\"\nfor candidate in \\\n    \"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)\u002Flmfdb-pipeline.conf\" \\\n    \"magma\u002Fscripts\u002Fdata-generation.conf\"; do\n  [ -f \"$candidate\" ] && { CONF=\"$candidate\"; break; }\ndone\nif [ -z \"$CONF\" ]; then\n  echo \"I'm sorry, I can't do that — no database pipeline conf found.\"\n  echo \"Run \u002Fconfigure-database (mathcity-lmfdb.configure-database) to create lmfdb-pipeline.conf at your project root.\"\n  echo \"(This conf holds your PostgreSQL and DATA_DIR settings for the LMFDB pipeline.)\"\n  exit 1\nfi\n```\n\nDo not design a type for a project with no pipeline.\n\n## Step 1 — Justification gate (LM4.1)\n\nAnswer, in writing, in the type's bead:\n\n1. **Distinct object class?** Is this a new class of mathematical\n   object, or a variant of an existing type? If the data fits as new\n   columns on an existing type → STOP, verdict **reject-as-type**, and\n   route to [[update-schema]] instead. (Test: would it share the\n   existing type's label scheme and storage path? If yes, it's\n   columns.)\n2. **Reused across sessions?** One-off experiment output is not a type\n   (LM2.5). If unsure → **defer**.\n3. **Populating intrinsics exist or are specified?** Name them with\n   signatures. If they don't exist yet → **defer** until specified.\n\n## Step 2 — Label scheme design (LM1.x, LM4.4) — BEFORE any code\n\nDraft the scheme against every LM1 rule:\n\n```\nLABEL SCHEME PROPOSAL — \u003Ctype name>\n\nComponents:   \u003Cordered list, most-significant first, with how each is computed>\nExample:      \u003Ca concrete label>\nDeterminism:  \u003Cwhy re-running assignment reproduces the label — LM1.2>\nCharset:      [0-9a-zA-Z.-] only; m-prefix negatives; base-62 hashes — LM1.3\nStructure:    \u003Cwhich components are invariants; any hash component and its\n               canonical-form justification — LM1.4>\nParent:       \u003Cparent type + prefix rule, or \"freestanding (no parent)\" — LM1.5>\nUniqueness:   \u003Cwhat disambiguates ties — ordinal, index, canonical sort — LM1.8>\n```\n\nModel on the existing schemes in `\u003Crepos-root>\u002Fhecke\u002Fmagma\u002Fschema.md`\n(\"Label Scheme\" section): order labels\n(`\u003Cdim>.\u003Cform_label>.\u003Cdisc_prefix>\u003Cabs_disc>...\u003CN>`), subgroup labels\n(`\u003Corder_label>.\u003Cindex>.\u003Cprelabel>.\u003Cordinal>`), eigenform labels\n(`\u003Csubgroup_label>-\u003Csuffix>`).\n\n## Step 3 — Present to the human adjudicator\n\nGate Steps 1–2 with [[present-it]] (compact form for a clean,\nprecedent-following scheme; full-form if the type opens a new label\nnamespace or a new storage layout):\n\n```\nDECISION:  Create LMFDB type \u003CLMFDBXxx> with label scheme \u003Cscheme>?\nCONTEXT:   \u003Cone sentence: what it stores, why not columns on an existing type>\nRECOMMEND: APPROVE — \u003Cone-line rationale>\nCONFIRM:   y \u002F n \u002F grill-me-further\n```\n\n**Do not write code before approval.** Record the approval per\ndecision-recording discipline. Also collect NOW (they gate later\nsteps):\n\n- storage layout: per-order vs per-entry, with the loading-pattern\n  rationale (LM4.5);\n- **webpage decision**: public-LMFDB-bound or internal-only (LM4.6);\n- final column list with types — changing them after the schema pass is\n  a data migration (LM4.4).\n\n## Step 4 — Scaffold\n\nInvoke [[create-lmfdb-type]] with the approved inputs. It drives the\n11-step scaffold (declaration + ATTRS through make scripts) and calls\n[[update-schema]] for the seven-file sync at its Step 10. Enforce as it\ngoes:\n\n- naming conventions LM4.2 (including `Sort([...])` +\n  `\u002F\u002FALWAYS ALPHABETICAL!`);\n- no `SaveJsonb()` on attribute assignment, no `.sig` edits (LM4.7);\n- every deferred checklist item (caching, make script, staged DB\n  wiring) gets a tracked issue\u002Fbead — untracked deferral blocks close\n  (LM4.3).\n\n## Step 5 — Webpage wiring (LM4.6)\n\n- **Public-bound:** run [[plan-an-lmfdb-webpage]] to produce the\n  web-layer task breakdown; link its output from the type's bead.\n- **Internal-only:** write \"internal-only, no webpage\" in the bead with\n  one sentence of rationale (precedent: `LMFDBOrderSnfs`).\n\n## Step 6 — Final gate\n\nRun [[check-lmfdb-hygiene]] on the type's bead\u002Fdiff. The type is done\nonly on **approve**. On **revise**, fix and re-run; on **reject**,\nescalate to the human adjudicator — do not close the bead. Then fire the documentation\nrules: README\u002F`schema.md` rows land with the change (update-README \u002F\nD-rules).\n\n## Hard rules\n\n- **Never skip Step 3.** No type is created without the human adjudicator's recorded\n  approval of the justification and label scheme.\n- **Label scheme freezes at approval.** Changing it after data exists is\n  an LM1.7 migration, not an edit.\n- **This skill does not push or commit.** Git operations follow the\n  active context profile and repo policy (`authorize-git-operation`\n  where required).\n- **reject-as-type is a success outcome.** Routing to `update-schema`\n  instead of creating a redundant type is the policy working.\n\n## Cross-references\n\n- [POLICY.md](..\u002F..\u002FPOLICY.md) — LM-rules; this skill executes LM4.1\u002FLM4.4\u002FLM4.6 and inherits the rest\n- [[create-lmfdb-type]] — the scaffold this skill wraps\n- [[update-schema]] — seven-file schema sync (and the destination for reject-as-type)\n- [[plan-an-lmfdb-webpage]] — Step 5 for public-bound types\n- [[check-lmfdb-hygiene]] — Step 6 final gate\n- [[new-hygiene-policy]] — the dev-subdomain sibling whose gate-on-the human adjudicator shape this follows\n- `\u003Crepos-root>\u002Fhecke\u002Fmagma\u002Fschema.md` — label-scheme precedents\n","# new-lmfdb-type-policy\n\nPolicy-gated wrapper around [[create-lmfdb-type]]. That skill scaffolds\nthe code; THIS skill makes sure the type deserves to exist, its label\nscheme is right before any code is written, and the result passes\n[POLICY.md](..\u002F..\u002FPOLICY.md) before the bead closes.\n\n**Why the wrapper exists:** labels are permanent (LM1.7) and a type is a\nseven-file contract (LM4.3). The expensive mistakes — wrong label\nscheme, type-that-should-have-been-columns, unrecorded webpage decision\n— all happen *before* `create-lmfdb-type`'s Step 1. This skill front-\nloads those decisions and gates them on the human adjudicator.\n\n## Step 0 — Read the policy and the dependency check\n\n```bash\ncat \u003Cmathcity-pack-root>\u002Fsubdomains\u002Flmfdb\u002FPOLICY.md\n```\n\nThen verify the pipeline conf exists before proceeding (P1.14):\n\n```bash\n# Discover conf: project root first, hecke fallback\nCONF=\"\"\nfor candidate in \\\n    \"$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)\u002Flmfdb-pipeline.conf\" \\\n    \"magma\u002Fscripts\u002Fdata-generation.conf\"; do\n  [ -f \"$candidate\" ] && { CONF=\"$candidate\"; break; }\ndone\nif [ -z \"$CONF\" ]; then\n  echo \"I'm sorry, I can't do that — no database pipeline conf found.\"\n  echo \"Run \u002Fconfigure-database (mathcity-lmfdb.configure-database) to create lmfdb-pipeline.conf at your project root.\"\n  echo \"(This conf holds your PostgreSQL and DATA_DIR settings for the LMFDB pipeline.)\"\n  exit 1\nfi\n```\n\nDo not design a type for a project with no pipeline.\n\n## Step 1 — Justification gate (LM4.1)\n\nAnswer, in writing, in the type's bead:\n\n1. **Distinct object class?** Is this a new class of mathematical\n   object, or a variant of an existing type? If the data fits as new\n   columns on an existing type → STOP, verdict **reject-as-type**, and\n   route to [[update-schema]] instead. (Test: would it share the\n   existing type's label scheme and storage path? If yes, it's\n   columns.)\n2. **Reused across sessions?** One-off experiment output is not a type\n   (LM2.5). If unsure → **defer**.\n3. **Populating intrinsics exist or are specified?** Name them with\n   signatures. If they don't exist yet → **defer** until specified.\n\n## Step 2 — Label scheme design (LM1.x, LM4.4) — BEFORE any code\n\nDraft the scheme against every LM1 rule:\n\n```\nLABEL SCHEME PROPOSAL — \u003Ctype name>\n\nComponents:   \u003Cordered list, most-significant first, with how each is computed>\nExample:      \u003Ca concrete label>\nDeterminism:  \u003Cwhy re-running assignment reproduces the label — LM1.2>\nCharset:      [0-9a-zA-Z.-] only; m-prefix negatives; base-62 hashes — LM1.3\nStructure:    \u003Cwhich components are invariants; any hash component and its\n               canonical-form justification — LM1.4>\nParent:       \u003Cparent type + prefix rule, or \"freestanding (no parent)\" — LM1.5>\nUniqueness:   \u003Cwhat disambiguates ties — ordinal, index, canonical sort — LM1.8>\n```\n\nModel on the existing schemes in `\u003Crepos-root>\u002Fhecke\u002Fmagma\u002Fschema.md`\n(\"Label Scheme\" section): order labels\n(`\u003Cdim>.\u003Cform_label>.\u003Cdisc_prefix>\u003Cabs_disc>...\u003CN>`), subgroup labels\n(`\u003Corder_label>.\u003Cindex>.\u003Cprelabel>.\u003Cordinal>`), eigenform labels\n(`\u003Csubgroup_label>-\u003Csuffix>`).\n\n## Step 3 — Present to the human adjudicator\n\nGate Steps 1–2 with [[present-it]] (compact form for a clean,\nprecedent-following scheme; full-form if the type opens a new label\nnamespace or a new storage layout):\n\n```\nDECISION:  Create LMFDB type \u003CLMFDBXxx> with label scheme \u003Cscheme>?\nCONTEXT:   \u003Cone sentence: what it stores, why not columns on an existing type>\nRECOMMEND: APPROVE — \u003Cone-line rationale>\nCONFIRM:   y \u002F n \u002F grill-me-further\n```\n\n**Do not write code before approval.** Record the approval per\ndecision-recording discipline. Also collect NOW (they gate later\nsteps):\n\n- storage layout: per-order vs per-entry, with the loading-pattern\n  rationale (LM4.5);\n- **webpage decision**: public-LMFDB-bound or internal-only (LM4.6);\n- final column list with types — changing them after the schema pass is\n  a data migration (LM4.4).\n\n## Step 4 — Scaffold\n\nInvoke [[create-lmfdb-type]] with the approved inputs. It drives the\n11-step scaffold (declaration + ATTRS through make scripts) and calls\n[[update-schema]] for the seven-file sync at its Step 10. Enforce as it\ngoes:\n\n- naming conventions LM4.2 (including `Sort([...])` +\n  `\u002F\u002FALWAYS ALPHABETICAL!`);\n- no `SaveJsonb()` on attribute assignment, no `.sig` edits (LM4.7);\n- every deferred checklist item (caching, make script, staged DB\n  wiring) gets a tracked issue\u002Fbead — untracked deferral blocks close\n  (LM4.3).\n\n## Step 5 — Webpage wiring (LM4.6)\n\n- **Public-bound:** run [[plan-an-lmfdb-webpage]] to produce the\n  web-layer task breakdown; link its output from the type's bead.\n- **Internal-only:** write \"internal-only, no webpage\" in the bead with\n  one sentence of rationale (precedent: `LMFDBOrderSnfs`).\n\n## Step 6 — Final gate\n\nRun [[check-lmfdb-hygiene]] on the type's bead\u002Fdiff. The type is done\nonly on **approve**. On **revise**, fix and re-run; on **reject**,\nescalate to the human adjudicator — do not close the bead. Then fire the documentation\nrules: README\u002F`schema.md` rows land with the change (update-README \u002F\nD-rules).\n\n## Hard rules\n\n- **Never skip Step 3.** No type is created without the human adjudicator's recorded\n  approval of the justification and label scheme.\n- **Label scheme freezes at approval.** Changing it after data exists is\n  an LM1.7 migration, not an edit.\n- **This skill does not push or commit.** Git operations follow the\n  active context profile and repo policy (`authorize-git-operation`\n  where required).\n- **reject-as-type is a success outcome.** Routing to `update-schema`\n  instead of creating a redundant type is the policy working.\n\n## Cross-references\n\n- [POLICY.md](..\u002F..\u002FPOLICY.md) — LM-rules; this skill executes LM4.1\u002FLM4.4\u002FLM4.6 and inherits the rest\n- [[create-lmfdb-type]] — the scaffold this skill wraps\n- [[update-schema]] — seven-file schema sync (and the destination for reject-as-type)\n- [[plan-an-lmfdb-webpage]] — Step 5 for public-bound types\n- [[check-lmfdb-hygiene]] — Step 6 final gate\n- [[new-hygiene-policy]] — the dev-subdomain sibling whose gate-on-the human adjudicator shape this follows\n- `\u003Crepos-root>\u002Fhecke\u002Fmagma\u002Fschema.md` — label-scheme precedents\n",{},"main","8d13efb5e66306aeae6e3b948bc463e7e2db1b65","skills\u002Fnew-lmfdb-type-policy","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public","https:\u002F\u002Fgithub.com\u002Ftdupu","https:\u002F\u002Ftdupu.github.io\u002Findex.html","Associate Professor, Department of Mathematics and Statistics, University of Vermont --- last updated Fall 2019","2026-11-13 09:07:19",{"source":7,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add tdupu\u002Fagent-skills-public --skill new-lmfdb-type-policy",{"tool":37,"command":38},"bunx","bunx skills add tdupu\u002Fagent-skills-public --skill new-lmfdb-type-policy",{"tool":40,"command":41},"pnpm","pnpm dlx skills add tdupu\u002Fagent-skills-public --skill new-lmfdb-type-policy","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fskills\u002Fnew-lmfdb-type-policy\u002FSKILL.md",true,[],[46,53,60,67,74,81],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":18},"01M7EW4PT586KKE7VRCAD80Z9K","writing-skills","Use when creating new skills, editing existing skills, or verifying skills work before deployment","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fsuperpowers\u002Fskills\u002Fwriting-skills\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":18},"01M7EW4HV5V74AF3PTV8N4T0PS","writing-plans","Use when you have a spec or requirements for a multi-step task, before touching code","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fsuperpowers\u002Fskills\u002Fwriting-plans\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":18},"01M7EW4DNKGP5CR1RAJP2J9PHA","verification-before-completion","Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fsuperpowers\u002Fskills\u002Fverification-before-completion\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":18},"01M7EW4BA2QP88VBFVXDVJM7XQ","using-superpowers","Use when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fsuperpowers\u002Fskills\u002Fusing-superpowers\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":18},"01M7EW47F2JG5FBESDJNP3PKMV","using-git-worktrees","Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fsuperpowers\u002Fskills\u002Fusing-git-worktrees\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":18},"01M7EW44N8556Y556NX9HJT4KP","test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002Ftdupu\u002Fagent-skills-public\u002Fblob\u002Fmain\u002Fsuperpowers\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[]]