[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75K7AVKFPCAKKX8BBK2HWWN":3,"skill-same:01M75K7AVKFPCAKKX8BBK2HWWN":49},{"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":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M75K7AVKFPCAKKX8BBK2HWWN","spec-walkthrough",null,"Render a spec bundle (or a chosen slice) into a plain-language, didactic comprehension artifact a human reads and judges for themselves: an unaided cold read before kickoff, re-orientation mid-execution, or onboarding to a finished or abandoned spec. Standalone and strictly read-only: it renders any status, never edits, signs off, mutates the pipeline, or delivers a verdict of its own. The independent complement to \u002Fspec-kickoff's guided dialogue.","inkatze\u002Fplanwright","inkatze","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F2029270?v=4","01M75K412KAPRFBSFXA7ZF2ZVY","Diego Romero","01M75K41364G95D0ZS5MWK7PZ3",4,"MIT","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Fskills\u002Fspec-walkthrough\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: spec-walkthrough\ndescription: >\n  Render a spec bundle (or a chosen slice) into a plain-language, didactic\n  comprehension artifact a human reads and judges for themselves: an unaided\n  cold read before kickoff, re-orientation mid-execution, or onboarding to a\n  finished or abandoned spec. Standalone and strictly read-only: it renders any\n  status, never edits, signs off, mutates the pipeline, or delivers a verdict of\n  its own. The independent complement to \u002Fspec-kickoff's guided dialogue.\nargument-hint: \"[--scope \u003Cselector>] [--reveal] \u003Cspec>\"\n---\n\n# \u002Fspec-walkthrough\n\nA standalone, read-only command that renders an existing spec bundle, or a\nselected part of it, into a visual, plain-English artifact a human reads and\n**judges for themselves** (REQ-A1.1, D-1). It lowers the cost of absorbing a\ndense bundle so a reader can independently evaluate it at any lifecycle stage,\nand it stands apart from `\u002Fspec-kickoff` precisely so that read stays\nindependent: the moment the authoring agent performs the comprehension,\nindependent review collapses into \"the agent reviewed its own spec.\"\n\nOn a successful load the command assembles the bundle into a single\nself-contained HTML file and writes it to the gitignored\n`.claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F` location (REQ-E1.1), naming the path in the load\nreport. The artifact carries the one-pager, the decision map, the drawn\ndependency graph, and the teach-back; a partial `--scope` selector renders only\nthe sections in its scope (a single file, a requirement group, the decision set,\nthe task graph, or one decision plus its blast radius), and the framing adapts\nto the bundle's auto-detected status (REQ-B1.2, REQ-B1.3).\n\n## Doctrine\n\nResolve and read these rule docs at run start via the rule-doc resolution\nconvention (`scripts\u002Fresolve-rule-doc.sh \u003Cdoc-name>`, or the documented\n`PLANWRIGHT_ROOT`\u002F`CLAUDE_PLUGIN_ROOT` chain): `spec-format` (the bundle this\ncommand reads conforms to the meta-spec — its file set, headers, and identifier\ndiscipline govern what the scaffold loads and validates) and `security-posture`\n(the identifier-charset and path-containment rules the scaffold enforces before\nany read, and the data-hygiene rule for anything it later writes). Their\ndefinitions govern wherever this skill names a concept. If one does not resolve,\nhalt with a clear message naming the missing doc and the chain consulted\n(REQ-K1.7: a clear message is the graceful arm; proceeding without doctrine is\nthe opaque failure).\n\nDoctrine manifest (the reading model above in machine-parseable form, per\n`doctrine\u002Finstruction-hygiene.md`; `run-start` loads before work begins):\n\nDoctrine: run-start spec-format\nDoctrine: run-start security-posture\n\n## Invocation\n\nRun from inside the work repository: the scaffold reads the bundle under the\nspec root `scripts\u002Fresolve-root.sh spec` resolves there, in any posture\n(`docs\u002Fspec-location.md`), and only reads. Invoke it as:\n\n```sh\nscripts\u002Fspec-walkthrough.sh [--scope \u003Cselector>] [--reveal] \u003Cspec>\n```\n\n- **`\u003Cspec>`** — the bare identifier, or its alias `specs\u002F\u003Cspec>`, with or\n  without one trailing slash. The identifier is charset-validated against\n  `^[a-z0-9][a-z0-9-]*$` (max 64) and the resolved path is containment-checked\n  **before any read** (REQ-A1.6). A hostile or malformed identifier or a path\n  that escapes the spec root is a clean refusal that never becomes a path and never\n  echoes the candidate back.\n- **`--scope \u003Cselector>`** — which part to render (REQ-A1.2, REQ-B1.2). Default\n  is the whole bundle. Selectors: `whole`; `file:\u003Cname>` (one of\n  `requirements`, `design`, `tasks`, `test-spec`); `reqs:\u003CGROUP>` (one\n  requirement group, e.g. `reqs:A`); `decisions` (the decision set); `tasks`\n  (the task graph); `decision:\u003Cid>` (a single decision plus its blast radius,\n  e.g. `decision:1` or `decision:D-1`).\n- **`--reveal`** — expose the underlying identifiers. Off by default (REQ-D1.3):\n  the default view is plain and audience-neutral.\n\n**Exit codes.** `0` the bundle loaded (full or partial); `1` graceful\ndegradation with nothing to load (the bundle directory is absent, holds none of\nthe four files, or the requested scope resolves to no part of it) — a clear\nmessage names what is absent or the available scopes, never an opaque halt\n(REQ-A1.5); `2` a clean refusal (a malformed invocation, or a hostile\u002Fmalformed\nidentifier or escaping path).\n\n**Artifact.** On a successful load the command assembles the bundle into a\nsingle self-contained HTML file (via `scripts\u002Fspec-assemble.sh`) and writes it\nto the gitignored `.claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F\u003Cspec>.html` (REQ-E1.1); the load\nreport names the path. The file opens offline in any browser with nothing\ninstalled (REQ-E1.2): the read (the one-pager) first, the teach-back prompt\nafter it, all bundle content HTML\u002FSVG-escaped so markup in the spec displays as\nliteral text and never executes (REQ-E1.7), identifiers behind an off-by-default\nreveal toggle (REQ-D1.3), original styling inlined that draws on MIT-licensed\ndesign primitives (REQ-E1.6), and a\nbundle + commit provenance stamp so a reader can tell whether it is stale\n(REQ-E1.5). Open it directly: `open .claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F\u003Cspec>.html`.\n\n**Optional Graphviz enhancement.** The dependency graph is always drawn as\ninline SVG (never ASCII), self-contained and offline. When the Graphviz `dot`\nbinary is present it is used for a richer node layout (read-only, for\ncoordinates only); when it is absent, exits non-zero, times out, or emits an\nunparseable layout, the view degrades identically to a built-in layout and\nrecords a one-line in-artifact note saying which path was taken (REQ-E1.3, D-5).\n`dot` is never on a path that can fail the render, so the artifact is fully\nself-contained and offline either way; only the layout differs. No installation\nis required: Graphviz is an enhancement, not a dependency. Two environment\nvariables tune the probe: `SPEC_WALKTHROUGH_DOT` overrides the binary name\n(default `dot`; point it at a name that does not exist to force the built-in\nlayout), and `SPEC_WALKTHROUGH_DOT_TIMEOUT` sets the watchdog in whole seconds\n(default 5; any value that is not a plain non-negative integer, such as `0.5`,\nis coerced back to 5) before a slow `dot` run is killed and the layout degrades.\n\n## Teach-back\n\nThe teach-back is the comprehension check: the reader restates the spec's own\nassertions, section by section, and judges each for themselves (D-3, D-9;\nREQ-C1.5). The claim set is extracted once by `scripts\u002Fspec-teachback.sh` over\nthe plain-language translation stream — every live requirement and every\ndecision becomes one claim, grouped into sections (one per requirement group,\nthen the decisions) — and **both** delivery paths render from that single\nsource, so they always cover the same claims:\n\n- **In-artifact** (assembled by the HTML task): a self-paced\n  agree \u002F disagree \u002F unsure checklist, one neutral choice per claim. It records\n  the reader's marks; it has no comparison engine and supplies no answer.\n- **In-session walk** (optional, this skill): present each claim section by\n  section, ask the reader to restate it in their own words, and record the\n  response. Where a restatement diverges from the source, **surface the\n  divergence** — show the verbatim source for the reader to compare — and let\n  the reader adjudicate (REQ-D1.4). Never say who is right, never score, never\n  supply the \"right\" answer: the independence firewall (REQ-D1.1) means the\n  tool presents and structures, the human judges.\n\nRun the full read first; only then offer the teach-back. The agent voices no\nverdict, score, or assessment of the spec at any point.\n\n## Invariants\n\nThese hold at every stage (contracts per `doctrine\u002Fhuman-gates.md`):\n\n- **Strictly read-only** (REQ-A1.3). The command never edits a bundle file,\n  commits, pushes, changes a status, or writes the kickoff brief. Its only\n  sanctioned write is the generated artifact to the gitignored\n  `.claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F` location; it modifies no tracked content (the\n  optional Graphviz probe's only writes are ephemeral, auto-cleaned `$TMPDIR`\n  temp files).\n- **Status-agnostic** (REQ-A1.4, REQ-B1.4). Every status renders — Draft, Ready,\n  Active, Done, Retired, Superseded — in deliberate contrast with the execution\n  skills' non-Active refusal. Rendering is read-only, so that refusal's safety\n  rationale does not apply, and terminal bundles stay valuable for archaeology.\n  There is no Active gate.\n- **Presents and structures; the human judges** (REQ-D1.1, the independence\n  firewall). The command never delivers a verdict, score, or quality assessment\n  of its own. It renders the bundle and records the reader's own responses; it\n  never supplies the \"right\" answer.\n- **Degrades, never halts opaquely** (REQ-A1.5). A missing, malformed, or\n  partial bundle, or an unresolvable scope, yields a clear message that names\n  what is present and what is absent.\n\n## Maintenance\n\nAfter each run, compare these instructions against the doctrine and spec they\nimplement: the `spec-format` and `security-posture` doctrine docs (the bundle\nfile set and identifier discipline this command reads, and the data-hygiene rule\nfor the artifact it writes) and REQ-F1.4 (the completion-time drift-observation\ncontract). If a concept this skill names (the bundle file set, the identifier\ncharset, the path-containment gate, or the artifact data-hygiene rule) has\nchanged meaning, gained or lost a step, or moved between docs, record a\none-line drift observation through the shared helper (`scripts\u002Fobs-record.sh\n--slug skill-drift --scope \u003Crepo> --text 'skill-drift(spec-walkthrough):\n\u003Cwhat>'` — the entry text keeps the `skill-drift(...)` prefix) and commit the\nfragment as its own chore commit, per REQ-B3.2 \u002F D-42; surface a non-zero\nhelper exit rather than silently dropping the observation. In repositories\nwithout `specs\u002F`, surface the drift to the user instead of recording it. Do\nnot edit this skill or the doctrine docs to resolve the drift; the\naccumulator's canonical reader (`\u002Fspec-draft`) owns folding drift into spec\namendments.\n","# \u002Fspec-walkthrough\n\nA standalone, read-only command that renders an existing spec bundle, or a\nselected part of it, into a visual, plain-English artifact a human reads and\n**judges for themselves** (REQ-A1.1, D-1). It lowers the cost of absorbing a\ndense bundle so a reader can independently evaluate it at any lifecycle stage,\nand it stands apart from `\u002Fspec-kickoff` precisely so that read stays\nindependent: the moment the authoring agent performs the comprehension,\nindependent review collapses into \"the agent reviewed its own spec.\"\n\nOn a successful load the command assembles the bundle into a single\nself-contained HTML file and writes it to the gitignored\n`.claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F` location (REQ-E1.1), naming the path in the load\nreport. The artifact carries the one-pager, the decision map, the drawn\ndependency graph, and the teach-back; a partial `--scope` selector renders only\nthe sections in its scope (a single file, a requirement group, the decision set,\nthe task graph, or one decision plus its blast radius), and the framing adapts\nto the bundle's auto-detected status (REQ-B1.2, REQ-B1.3).\n\n## Doctrine\n\nResolve and read these rule docs at run start via the rule-doc resolution\nconvention (`scripts\u002Fresolve-rule-doc.sh \u003Cdoc-name>`, or the documented\n`PLANWRIGHT_ROOT`\u002F`CLAUDE_PLUGIN_ROOT` chain): `spec-format` (the bundle this\ncommand reads conforms to the meta-spec — its file set, headers, and identifier\ndiscipline govern what the scaffold loads and validates) and `security-posture`\n(the identifier-charset and path-containment rules the scaffold enforces before\nany read, and the data-hygiene rule for anything it later writes). Their\ndefinitions govern wherever this skill names a concept. If one does not resolve,\nhalt with a clear message naming the missing doc and the chain consulted\n(REQ-K1.7: a clear message is the graceful arm; proceeding without doctrine is\nthe opaque failure).\n\nDoctrine manifest (the reading model above in machine-parseable form, per\n`doctrine\u002Finstruction-hygiene.md`; `run-start` loads before work begins):\n\nDoctrine: run-start spec-format\nDoctrine: run-start security-posture\n\n## Invocation\n\nRun from inside the work repository: the scaffold reads the bundle under the\nspec root `scripts\u002Fresolve-root.sh spec` resolves there, in any posture\n(`docs\u002Fspec-location.md`), and only reads. Invoke it as:\n\n```sh\nscripts\u002Fspec-walkthrough.sh [--scope \u003Cselector>] [--reveal] \u003Cspec>\n```\n\n- **`\u003Cspec>`** — the bare identifier, or its alias `specs\u002F\u003Cspec>`, with or\n  without one trailing slash. The identifier is charset-validated against\n  `^[a-z0-9][a-z0-9-]*$` (max 64) and the resolved path is containment-checked\n  **before any read** (REQ-A1.6). A hostile or malformed identifier or a path\n  that escapes the spec root is a clean refusal that never becomes a path and never\n  echoes the candidate back.\n- **`--scope \u003Cselector>`** — which part to render (REQ-A1.2, REQ-B1.2). Default\n  is the whole bundle. Selectors: `whole`; `file:\u003Cname>` (one of\n  `requirements`, `design`, `tasks`, `test-spec`); `reqs:\u003CGROUP>` (one\n  requirement group, e.g. `reqs:A`); `decisions` (the decision set); `tasks`\n  (the task graph); `decision:\u003Cid>` (a single decision plus its blast radius,\n  e.g. `decision:1` or `decision:D-1`).\n- **`--reveal`** — expose the underlying identifiers. Off by default (REQ-D1.3):\n  the default view is plain and audience-neutral.\n\n**Exit codes.** `0` the bundle loaded (full or partial); `1` graceful\ndegradation with nothing to load (the bundle directory is absent, holds none of\nthe four files, or the requested scope resolves to no part of it) — a clear\nmessage names what is absent or the available scopes, never an opaque halt\n(REQ-A1.5); `2` a clean refusal (a malformed invocation, or a hostile\u002Fmalformed\nidentifier or escaping path).\n\n**Artifact.** On a successful load the command assembles the bundle into a\nsingle self-contained HTML file (via `scripts\u002Fspec-assemble.sh`) and writes it\nto the gitignored `.claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F\u003Cspec>.html` (REQ-E1.1); the load\nreport names the path. The file opens offline in any browser with nothing\ninstalled (REQ-E1.2): the read (the one-pager) first, the teach-back prompt\nafter it, all bundle content HTML\u002FSVG-escaped so markup in the spec displays as\nliteral text and never executes (REQ-E1.7), identifiers behind an off-by-default\nreveal toggle (REQ-D1.3), original styling inlined that draws on MIT-licensed\ndesign primitives (REQ-E1.6), and a\nbundle + commit provenance stamp so a reader can tell whether it is stale\n(REQ-E1.5). Open it directly: `open .claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F\u003Cspec>.html`.\n\n**Optional Graphviz enhancement.** The dependency graph is always drawn as\ninline SVG (never ASCII), self-contained and offline. When the Graphviz `dot`\nbinary is present it is used for a richer node layout (read-only, for\ncoordinates only); when it is absent, exits non-zero, times out, or emits an\nunparseable layout, the view degrades identically to a built-in layout and\nrecords a one-line in-artifact note saying which path was taken (REQ-E1.3, D-5).\n`dot` is never on a path that can fail the render, so the artifact is fully\nself-contained and offline either way; only the layout differs. No installation\nis required: Graphviz is an enhancement, not a dependency. Two environment\nvariables tune the probe: `SPEC_WALKTHROUGH_DOT` overrides the binary name\n(default `dot`; point it at a name that does not exist to force the built-in\nlayout), and `SPEC_WALKTHROUGH_DOT_TIMEOUT` sets the watchdog in whole seconds\n(default 5; any value that is not a plain non-negative integer, such as `0.5`,\nis coerced back to 5) before a slow `dot` run is killed and the layout degrades.\n\n## Teach-back\n\nThe teach-back is the comprehension check: the reader restates the spec's own\nassertions, section by section, and judges each for themselves (D-3, D-9;\nREQ-C1.5). The claim set is extracted once by `scripts\u002Fspec-teachback.sh` over\nthe plain-language translation stream — every live requirement and every\ndecision becomes one claim, grouped into sections (one per requirement group,\nthen the decisions) — and **both** delivery paths render from that single\nsource, so they always cover the same claims:\n\n- **In-artifact** (assembled by the HTML task): a self-paced\n  agree \u002F disagree \u002F unsure checklist, one neutral choice per claim. It records\n  the reader's marks; it has no comparison engine and supplies no answer.\n- **In-session walk** (optional, this skill): present each claim section by\n  section, ask the reader to restate it in their own words, and record the\n  response. Where a restatement diverges from the source, **surface the\n  divergence** — show the verbatim source for the reader to compare — and let\n  the reader adjudicate (REQ-D1.4). Never say who is right, never score, never\n  supply the \"right\" answer: the independence firewall (REQ-D1.1) means the\n  tool presents and structures, the human judges.\n\nRun the full read first; only then offer the teach-back. The agent voices no\nverdict, score, or assessment of the spec at any point.\n\n## Invariants\n\nThese hold at every stage (contracts per `doctrine\u002Fhuman-gates.md`):\n\n- **Strictly read-only** (REQ-A1.3). The command never edits a bundle file,\n  commits, pushes, changes a status, or writes the kickoff brief. Its only\n  sanctioned write is the generated artifact to the gitignored\n  `.claude\u002Fwalkthroughs\u002F\u003Cspec>\u002F` location; it modifies no tracked content (the\n  optional Graphviz probe's only writes are ephemeral, auto-cleaned `$TMPDIR`\n  temp files).\n- **Status-agnostic** (REQ-A1.4, REQ-B1.4). Every status renders — Draft, Ready,\n  Active, Done, Retired, Superseded — in deliberate contrast with the execution\n  skills' non-Active refusal. Rendering is read-only, so that refusal's safety\n  rationale does not apply, and terminal bundles stay valuable for archaeology.\n  There is no Active gate.\n- **Presents and structures; the human judges** (REQ-D1.1, the independence\n  firewall). The command never delivers a verdict, score, or quality assessment\n  of its own. It renders the bundle and records the reader's own responses; it\n  never supplies the \"right\" answer.\n- **Degrades, never halts opaquely** (REQ-A1.5). A missing, malformed, or\n  partial bundle, or an unresolvable scope, yields a clear message that names\n  what is present and what is absent.\n\n## Maintenance\n\nAfter each run, compare these instructions against the doctrine and spec they\nimplement: the `spec-format` and `security-posture` doctrine docs (the bundle\nfile set and identifier discipline this command reads, and the data-hygiene rule\nfor the artifact it writes) and REQ-F1.4 (the completion-time drift-observation\ncontract). If a concept this skill names (the bundle file set, the identifier\ncharset, the path-containment gate, or the artifact data-hygiene rule) has\nchanged meaning, gained or lost a step, or moved between docs, record a\none-line drift observation through the shared helper (`scripts\u002Fobs-record.sh\n--slug skill-drift --scope \u003Crepo> --text 'skill-drift(spec-walkthrough):\n\u003Cwhat>'` — the entry text keeps the `skill-drift(...)` prefix) and commit the\nfragment as its own chore commit, per REQ-B3.2 \u002F D-42; surface a non-zero\nhelper exit rather than silently dropping the observation. In repositories\nwithout `specs\u002F`, surface the drift to the user instead of recording it. Do\nnot edit this skill or the doctrine docs to resolve the drift; the\naccumulator's canonical reader (`\u002Fspec-draft`) owns folding drift into spec\namendments.\n",{"name":5,"description":24,"argument-hint":25},"Render a spec bundle (or a chosen slice) into a plain-language, didactic comprehension artifact a human reads and judges for themselves: an unaided cold read before kickoff, re-orientation mid-execution, or onboarding to a finished or abandoned spec. Standalone and strictly read-only: it renders any status, never edits, signs off, mutates the pipeline, or delivers a verdict of its own. The independent complement to \u002Fspec-kickoff's guided dialogue.\n","[--scope \u003Cselector>] [--reveal] \u003Cspec>","main","4e6ff2890011c9b5c4cf4ce33e05a037cc8996a6","skills\u002Fspec-walkthrough",0,"https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright","https:\u002F\u002Fgithub.com\u002Finkatze","https:\u002F\u002Fwww.inkatze.com","Software Engineer | Open Source n00b","2026-11-15 05:06:45",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add inkatze\u002Fplanwright --skill spec-walkthrough",{"tool":41,"command":42},"bunx","bunx skills add inkatze\u002Fplanwright --skill spec-walkthrough",{"tool":44,"command":45},"pnpm","pnpm dlx skills add inkatze\u002Fplanwright --skill spec-walkthrough","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Fskills\u002Fspec-walkthrough\u002FSKILL.md",true,[],[50,56,62,67,74,81],{"skillNo":51,"name":52,"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":53,"categories":54,"keywords":55,"publishedAt":19},"01M75K7QKF8AFQ3JAJ1T67HT3F","demo","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fscript-invocation\u002Fsh-runs-bash\u002Fskills\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":57,"name":58,"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":59,"categories":60,"keywords":61,"publishedAt":19},"01M75K7NKAW0V15FAVSWPVV4NZ","usage","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fscript-invocation\u002Fclean\u002Fskills\u002Fusage\u002FSKILL.md",[],[],{"skillNo":63,"name":52,"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":64,"categories":65,"keywords":66,"publishedAt":19},"01M75K7MD8DA50HDQ1NSTZCVXM","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fscript-invocation\u002Fclean\u002Fskills\u002Fdemo\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M75K7GFAKYN7X3QZ385P3V7H","tower","The front door: a standing conversational session that takes any ask about the repository in plain language, answers questions, and routes work that changes the repository per request onto visual flight (an isolated worker that lands a draft PR) or instrument flight (the spec pipeline), stating its grounds, with status on demand. Use it to start work without writing a spec first. It never edits the repository itself, never merges, never marks a PR ready, and never starts a kickoff.","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Fskills\u002Ftower\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"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},"01M75K72A3397KH8GXFDJ0FCJM","spec-draft","Interactively elicit a four-file spec bundle (requirements.md, design.md, tasks.md, test-spec.md) at Status Draft on the spec's own branch, mining seed sources (pending notes, the observations log, transcripts) and citing them. Runs fold-detection against existing specs on every invocation and surfaces an extend recommendation instead of spinning a duplicate bundle. Commits the completed bundle (commit_on_draft opt-out); never pushes, never flips a spec to Ready.","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Fskills\u002Fspec-draft\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M75K6XYRVCS69Z79MF4N71GV","self-review","Gate-wired review pass of the current feature branch: Discovery Rigor fan-out, multi-pass validation, and act-then-review dispositions per the finding categorization. Standalone runs push and open or update a draft PR; pass --nested to stay local and return the audit record to the invoking skill.","https:\u002F\u002Fgithub.com\u002Finkatze\u002Fplanwright\u002Fblob\u002Fmain\u002Fskills\u002Fself-review\u002FSKILL.md",[],[]]