[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M752S7PYMDGF1KT0HCNG8D15":3,"skill-same:01M752S7PYMDGF1KT0HCNG8D15":42},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":14,"categories":15,"keywords":16,"publishedAt":17,"linkOnly":18,"markdownContent":19,"markdownBody":19,"frontMatter":20,"refName":21,"commitSha":22,"skillDirectory":23,"forksCount":24,"repositoryHtmlUrl":25,"repositoryHomepageUrl":6,"ownerHtmlUrl":26,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":27,"install":28,"resources":41},"01M752S7PYMDGF1KT0HCNG8D15","sd-atomic-task-format",null,"developzoneio\u002Fspecwright","developzoneio","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F288244965?v=4","01M752S4GQXBCH6H67H28M9HTN","01M752S4HXZ73BTCHG2K1VCYQA",10,"MIT","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Fskills\u002Fsd-atomic-task-format\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"# sd-atomic-task-format\n\nCanonical format for atomic implementation tasks in specwright.\nUsed by `sd-spec-architect` when authoring `02-tasks.md` and by `sd-implementer` when reading task contracts.\n\n---\n\n## Task block (11 required fields)\n\n```markdown\n### T\u003CNN> - \u003Cimperative title>\n\n- **Files**: \u003Ccomma-separated relative paths>\n- **Layer**: \u003CDomain | Application | Infrastructure | Presentation | Tests | Config>\n- **Step type**: \u003Cfoundation | behavior | wiring | polish | test>\n- **Test**: \u003Cwhich test file(s) cover this task>\n- **Acceptance**: \u003Cobservable criterion>\n- **Covers**: \u003CSC-\u002FAC-ID list, e.g. SC-1, AC-2 | none>\n- **Depends on**: \u003CT## | none>\n- **Conflicts with**: \u003CT## list | none>\n- **Estimated complexity**: \u003CS | M | L>\n- **Reversibility**: \u003Ctrivial | moderate | hard>\n- **Pattern refs**: \u003C1-3 file:line precedent citations + what to mirror | none>\n- **Status**: open\n```\n\nThe last line, `Status`, is the **check-off marker** - execution state, not part of the task's\ncontract. See \"Check-off marker\" below; it is not counted among the 11 required fields.\n\nAll 11 fields are **required**, not optional. A task block missing any of them is malformed.\n\n`Pattern refs` is required on **every** task - the old \"only when the task creates a new file or a\nnew public symbol\" condition is gone. A task with no precedent worth citing writes\n`Pattern refs: none` **explicitly**. The point of the field is the assertion: `none` says the\narchitect looked and found nothing, while an absent field says nothing at all, and the two are not\nthe same claim.\n\nReading legacy specs stays lenient. A block authored before this rule, with the field absent, is\nread as `Pattern refs: none` so existing `.specs\u002F` folders keep working. `\u002Fsd:spec validate`\nreports the omission as `SL060` (WARN) - it does not block, and nothing downstream refuses the\ntask.\n\n### Refactor mode adds one field\n\n`\u002Fsd:refactor` tasks append one more field after `Pattern refs`:\n\n```markdown\n- **Parallel batch**: \u003Cbatch number | \"solo\">\n```\n\nTasks sharing a batch number have disjoint file sets and no `Depends on` \u002F `Conflicts with`\nrelationship between them - they are safe to execute in parallel. `solo` means the task cannot\nbe batched with any other. Other workflow types (feature, bug, perf) do not use this field.\n\n### Port mode constrains two fields\n\n`\u002Fsd:port` tasks (authored by `sd-spec-architect` against a `PORT-` spec) narrow two of the 11\nrequired fields rather than adding a new one:\n\n- **`Pattern refs`** is never `none`. Every port task cites 1-3 snapshot member ranges in the\n  shape `04-artifacts\u002Fsource\u002F\u003Cpath>:\u003Cfirst>-\u003Clast>`, never a host sibling path and never prose -\n  the citation IS the anti-drift precedent, not a suggestion of one.\n- **`Acceptance`** carries a `Licensed deviations:` line naming the deviation-table `D\u003CNN>` IDs\n  that apply to this task, or `Licensed deviations: none` when no deviation applies - explicit\n  `none` is compliant, an absent line is not, same convention `Pattern refs: none` already uses.\n\n```markdown\n- **Pattern refs**: 04-artifacts\u002Fsource\u002Forders\u002Forder-intake:12-58 - mirror validation order\n- **Acceptance**: CheckTotals runs before the ledger write. Licensed deviations: D02\n```\n\n`\u002Fsd:spec validate` checks both (`SL061`-`SL066`) against `04-artifacts\u002Fsource\u002FMANIFEST.md` and the\nspec's `## Deviation table` - see `commands\u002Fspec.md`'s \"Port task-block checks\".\n\n### Re-plan adds one field\n\nA task **regenerated by a mid-execution re-plan** (`\u002Fsd:feature`, `\u002Fsd:refactor` - see the\n**sd-replan-loop** skill) carries one extra field after `Pattern refs` (or after `Parallel batch`\nin refactor mode):\n\n```markdown\n- **Revised-by**: \u003CR## - the revision entry in 01-plan.md that regenerated this task>\n```\n\n`Revised-by` is **conditional, not one of the 11 required fields**: it is present only on a task the\nre-plan gate regenerated, and absent on every task authored in the original Plan phase. It ties the\ntask to its `## Revisions` entry in `01-plan.md`; `\u002Fsd:spec validate` checks the two agree in both\ndirections (`SL070`-`SL073`). A task authored at Plan phase never carries it - do not add it\nspeculatively.\n\n---\n\n## Check-off marker\n\n`\u002Fsd:feature`, `\u002Fsd:port` and `\u002Fsd:refactor` resume from whether a task is **checked**. This\nsection is the only definition of what that means - commands reference it and never restate it.\n\n### Canonical form\n\n```markdown\n- **Status**: \u003Copen | done>\n```\n\n- It is the **last line** of the block - after `Pattern refs`, and after `Parallel batch` \u002F\n  `Revised-by` when those are present.\n- `sd-spec-architect` writes `- **Status**: open` on **every** task it authors, including tasks a\n  re-plan regenerates. \"Unchecked\" is then an assertion on the page, not an absence.\n- The main thread **checks off** a task by rewriting that one line to `- **Status**: done`. Nothing\n  else in the block changes. `sd-implementer` never edits `02-tasks.md`.\n- Unlike the 11 contract fields, `Status` is the one value that changes after planning. It records\n  execution state, so it is not one of the 11 required fields.\n\n**Why a field and not a heading prefix** (`### [x] T01 - ...`, the form a model picks unprompted):\nthe heading is the task's identity. `T\u003CNN>` is cited by `Depends on`, `Conflicts with`,\n`Revised-by`, `\u002Fsd:spec validate` findings, `05-retro.md` lines and `\u002Fsd:status` counters.\nRewriting the heading on check-off is the exact drift ADR 0002 recorded: `### ✅ T01` read as zero\ntasks by a naive counter. A field keeps the heading stable and reuses the field label grammar\nbelow, so there is no second parser. It also has a written-out unchecked state (`open`), while a\nheading has none. And a `[x]` in a heading is not rendered as a checkbox anyway - GitHub-flavored\nMarkdown renders task-list boxes only on list items.\n\n### Reading\n\nMatch the label and value with the field label grammar below; the value is case-insensitive.\nResolve each task in this order:\n\n| Block carries | Reads as |\n|---|---|\n| `Status: done` | checked |\n| `Status: open` | unchecked |\n| `Status` with any other value | unchecked, reported as `SL067` |\n| No `Status`, heading has `[x]` or `✅` before the `T\u003CNN>` token | checked (drift), reported as `SL067` |\n| No `Status`, no heading prefix | unchecked (legacy) |\n\nWhen `Status` is present it decides - a heading prefix never overrides it. A heading prefix is\nhonoured only as a fallback so a run that checked off in a drifted form is not re-executed.\n\n**Spec status decides first.** Task markers are read **only while the spec's frontmatter status\nis `in-progress`**. A `done` or `archived` spec resolves from its frontmatter before any task row in\na resume state machine is evaluated. So a legacy spec with no markers at all (authored before this\nsection existed) and status `done` - e.g. `examples\u002Ffixture-project\u002F.specs\u002FFEAT-todo-priority` - is\nnever resumed into task execution, even though every one of its tasks reads as unchecked.\n\n`\u002Fsd:spec validate` reports a non-canonical marker as `SL067` (WARN). An absent `Status` is **not**\na finding: legacy blocks read as unchecked and the rule above keeps finished specs out of resume.\n\n---\n\n## Field label grammar\n\nThe block above shows the **canonical form to author**. It is an example, not the parsing rule -\nreal specs have drifted, and a reader that accepts only the canonical form rejects valid work.\n\n**When authoring**, always emit the canonical form:\n\n```markdown\n- **Files**: \u003Cvalue>\n```\n\n**When reading**, match tolerantly. A field label matches when all of the following hold:\n\n| Aspect | Rule |\n|---|---|\n| Bullet marker | `-` or `*`, any leading indentation |\n| Emphasis | `**` around the label is optional |\n| Colon | may sit inside the emphasis (`**Files:**`) or outside it (`**Files**:`) |\n| Case | label match is case-insensitive |\n| Whitespace | any amount around the marker, label, colon, and value |\n\nAll three of these are the same field and must parse identically:\n\n```markdown\n- **Files**: src\u002FFoo.cs\n- Files: src\u002FFoo.cs\n- **Files:** src\u002FFoo.cs\n```\n\n**Value extent.** A field's value runs from after the colon to the start of the next field label\nor the end of the block - it is **not** limited to one line. Indented continuation lines and\nnested sub-bullets belong to the field above them. `Acceptance` and `Pattern refs` are routinely\nauthored as multi-line values with nested bullets; a line-oriented reader that stops at the first\nnewline truncates them.\n\nThis grammar applies to **every** field in the block, not only to the field a given check cares\nabout. Anything that reads `02-tasks.md` - `\u002Fsd:spec validate`, `\u002Fsd:verify`, `sd-implementer` -\nuses this one rule. Do not write a per-field matcher.\n\n---\n\n## Field rules\n\n### Files\nMust exist (for edits) or be a new file path consistent with project conventions. Use `Glob` to verify before writing.\n\n### Layer\nMust come from the constitution's declared layers. Do not invent layer names. If a layer is not in `constitution.md`, flag as an Open question.\n\n### Step type\n| Value | Meaning |\n|---|---|\n| `foundation` | New abstractions, types, interfaces, DB schema — no behavior yet |\n| `behavior` | Business logic, event handlers, domain rules |\n| `wiring` | DI registration, configuration, middleware, glue code |\n| `polish` | Naming, comments, docs, dead-code removal |\n| `test` | Test files only — no production code change |\n\n### Acceptance\nMust be **observable**: a passing test, a 201 response, a method called exactly once. \"Feels right\" or \"code is cleaner\" are not acceptable. Every acceptance criterion must be verifiable without running the full app (unit\u002Fintegration test preferred). On a port task, also carries a `Licensed deviations:` line - see \"Port mode constrains two fields\" below.\n\n### Covers\n\nComma-separated scenario (SC-\u003Cn>) and success-criterion (AC-\u003Cn>) IDs from `00-spec.md` that\nthis task implements or proves. `none` is allowed only for pure wiring\u002Fpolish tasks that\nadvance no criterion directly. Every ID referenced must exist in the spec; every SC and AC in\nthe spec must be covered by at least one task - `\u002Fsd:verify` fails the spec otherwise. Specs\nauthored before this field existed (no SC\u002FAC IDs) are handled by `\u002Fsd:verify`'s generic\nchecks; treat a missing field as `Covers: none` when reading legacy `02-tasks.md` files.\nSpec types whose templates carry no SC-\u002FAC-IDs (refactor, bug, perf, rca) use `Covers: none`\nfor every task - the coverage requirement (VF010\u002FVF011) applies only to specs whose\n`00-spec.md` declares SC-\u002FAC-IDs, currently feature specs.\n\n### Estimated complexity\n| Value | Guideline |\n|---|---|\n| `S` | ≤ 30 lines changed, single file, single concept |\n| `M` | 30–100 lines, up to 3 files, one concern crossing a boundary |\n| `L` | > 100 lines OR multiple files OR requires design decision at implementation time |\n\n### Reversibility\n| Value | Meaning |\n|---|---|\n| `trivial` | Can be undone with a single revert \u002F delete |\n| `moderate` | Requires a few file reverts; no downstream migrations |\n| `hard` | Data migration, public API change, or multi-service impact |\n\n### Depends on \u002F Conflicts with\nDrive sequencing and batch planning. Tasks that **conflict** cannot run in the same parallel batch. Tasks that **depend on** a prior task cannot start until that task's acceptance criterion is met.\n\n### Pattern refs\n1-3 `file:line` citations of precedent code the implementer reads BEFORE writing, each with a one-line instruction of what to mirror (e.g. \"mirror handler structure and registration\", \"reuse this helper - do not duplicate\"). Required on every task; write `none` explicitly when there is genuinely no precedent to cite, which should be rare outside pure polish work. Verify each cited file exists before writing the ref. Discovery and adherence rules live in the **sd-pattern-discipline** skill.\n\n---\n\n## Atomicity rules\n\n- **One concern per task.** If the title contains \"and\" connecting two distinct concerns, split into two tasks.\n- **Single-file preferred.** Multi-file tasks should have a clear reason (e.g. interface + implementation pair).\n- If a task feels like two changes during implementation, STOP and surface to the main thread.\n\n---\n\n## Anti-patterns\n\n- Bundling interface + behavior + test into a single task — these are three tasks.\n- Using vague acceptance criteria (\"works correctly\", \"looks good\").\n- Listing files you haven't verified exist (for edits) or naming conventions don't match.\n- Inventing layer names not declared in the constitution.\n- Leaving `Depends on` \u002F `Conflicts with` empty when sequential or conflicting relationships exist.\n- Authoring a new-file task with `Pattern refs: none` - the implementer has no precedent to mirror.\n- Citing Pattern refs you did not verify exist.\n- Omitting `Pattern refs` instead of writing `none`. Silence is not an assertion; `SL060` flags it.\n- Checking off a task by editing its heading (`### [x] T01`, `### ✅ T01`) or appending a list\n  checkbox. The only check-off is `Status: done` - see \"Check-off marker\". `SL067` flags the rest.\n- Burying a precedent in prose beside the block instead of putting it in the field. `sd-implementer`\n  reads `TASK_DETAILS`, not the surrounding narrative - a ref outside the field does not reach it.\n",{},"main","cb6a2aef549bb625b9ba1869d220d40938e83c18","skills\u002Fsd-atomic-task-format",0,"https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio","2026-11-11 17:12:43",{"source":7,"skillName":5,"commands":29,"prompt":39,"zipAvailable":40},[30,33,36],{"tool":31,"command":32},"npx","npx skills add developzoneio\u002Fspecwright --skill sd-atomic-task-format",{"tool":34,"command":35},"bunx","bunx skills add developzoneio\u002Fspecwright --skill sd-atomic-task-format",{"tool":37,"command":38},"pnpm","pnpm dlx skills add developzoneio\u002Fspecwright --skill sd-atomic-task-format","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Fskills\u002Fsd-atomic-task-format\u002FSKILL.md",true,[],[43,49,54,60,65,70],{"skillNo":44,"name":45,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":46,"categories":47,"keywords":48,"publishedAt":17},"01M752V2C4FNMKBZTZETKKN70B","sd-model-escalation","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Ftests\u002Fcontract-lint\u002Ffixtures\u002Ffp-cl605-alias-list\u002Foverlay\u002Fskills\u002Fsd-model-escalation\u002FSKILL.md",[],[],{"skillNo":50,"name":45,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":17},"01M752V0C3S0BHSXZ4JM3B7MF2","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Ftests\u002Fcontract-lint\u002Ffixtures\u002Ffp-cl601-allow-comment\u002Foverlay\u002Fskills\u002Fsd-model-escalation\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":17},"01M752TYQZVASA2BTDRZB4N8W6","sd-demo-guard","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Ftests\u002Fcontract-lint\u002Ffixtures\u002Ffp-cl009-phrase-outside-phase0\u002Foverlay\u002Fskills\u002Fsd-demo-guard\u002FSKILL.md",[],[],{"skillNo":61,"name":45,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":17},"01M752TXBFQHK7NCDTZEBHZRFN","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Ftests\u002Fcontract-lint\u002Ffixtures\u002Fcl605-command-restates-ladder\u002Foverlay\u002Fskills\u002Fsd-model-escalation\u002FSKILL.md",[],[],{"skillNo":66,"name":45,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":17},"01M752TVDG89BJRTGFET80RCK2","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Ftests\u002Fcontract-lint\u002Ffixtures\u002Fcl604-row-skips-rung\u002Foverlay\u002Fskills\u002Fsd-model-escalation\u002FSKILL.md",[],[],{"skillNo":71,"name":45,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":8,"repositoryNo":11,"starsCount":12,"licenseSpdxId":13,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":17},"01M752TT6SD0SSK6Z0GQ4HM68P","https:\u002F\u002Fgithub.com\u002Fdevelopzoneio\u002Fspecwright\u002Fblob\u002Fmain\u002Ftests\u002Fcontract-lint\u002Ffixtures\u002Fcl603-full-model-id-in-row\u002Foverlay\u002Fskills\u002Fsd-model-escalation\u002FSKILL.md",[],[]]