[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6KKXGEPMX39R704WNK1VB7B":3,"skill-same:01M6KKXGEPMX39R704WNK1VB7B":100},{"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":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":48,"repositoryHtmlUrl":49,"repositoryHomepageUrl":6,"ownerHtmlUrl":50,"ownerBlogUrl":6,"ownerDescription":51,"updatedAt":52,"install":53,"resources":65},"01M6KKXGEPMX39R704WNK1VB7B","objective-create",null,"Create a new Objective record under .ns\u002Fobjectives\u002F\u003Cslug>\u002F. Use when starting to track a new unit of planned work: 'create an objective', 'turn this into a tracked objective', or scaffolding a new objective's thesis\u002Fscope\u002Froadmap. To act on an Objective that already exists, use objective-next (recommend work), objective-update (record progress), or objective-close (finish it).","nseng-ai\u002Fns","nseng-ai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F273792438?v=4","01M6KKN105A07WPV2TZKZXZGQF","nseng.ai","01M6KKN14MDSR9VF9739AWXG8M",2,"MIT","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"objective create","create","act","track","recommend","true","it","unit","close","scope","progress","thesis","scaffolding","next","record","finish","roadmap","use","work","update","2026-11-10 10:12:30",false,"---\nname: objective-create\ndisable-model-invocation: true\ndescription: \"Create a new Objective record under .ns\u002Fobjectives\u002F\u003Cslug>\u002F. Use when starting to track a new unit of planned work: 'create an objective', 'turn this into a tracked objective', or scaffolding a new objective's thesis\u002Fscope\u002Froadmap. To act on an Objective that already exists, use objective-next (recommend work), objective-update (record progress), or objective-close (finish it).\"\n---\n\n# objective-create\n\nCreate exactly one new Objective record under `.ns\u002Fobjectives\u002F\u003Cslug>\u002F`. Use the `objective` umbrella skill first for shared vocabulary, selection rules, storage model, safety boundaries, and family policy.\n\nThe Objective pattern set — wayfinding\u002Fideation, steelthread, standing, umbrella, autoobjective, and readme-driven-development — lives in `references\u002F\u003Cpattern>-create.md`; each reference owns that pattern's creation procedure on top of this skill's record mechanics. The pattern is chosen in the interview (see the Pattern beat below). When the user names a pattern upfront, read the matching reference immediately and skip the menu. Recognition prose and composition facts stay in the `objective` skill's patterns catalog (`references\u002Fobjective-patterns.md`).\n\n## Required shape\n\nNew records live in the active root only, `.ns\u002Fobjectives\u002F\u003Cslug>\u002F`, with `objective.md`, `roadmap.md`, and `updates\u002F` per the umbrella skill's storage model. Creation deltas: `orientation.md` is optional and orienting-only; there is no `closed.md` at creation.\n\n`objective.md` uses the umbrella skill's required headings.\n\n- `## Assumptions and Risks` must distinguish assumptions from risks, with enough context for a future `objective-update` to mark an assumption incorrect, a risk de-risked or not, or add newly discovered ones.\n- When the record accepts a deliberate shortcut — a prototype-race sacrifice of security, scope, or quality — never record the shortcut alone: name its upgrade beside it, so the later swap is a slice rather than a redesign and a future `objective-update` can retire the pair together.\n- Default to planning-only: execution-policy sections are never mandatory and appear only when the user explicitly asks for execution-friendly\u002Frunner\u002Fautonomous behavior. Never offer execution policy as an interview question on your own initiative — the Pattern beat settles this axis, and choosing a standard Objective (or any pattern other than autoobjective) without an explicit execution request means planning-only; execution policy can always be added later via `objective-update`. When the user does ask, read `references\u002Fexecution-friendly-create.md` before asking policy questions.\n- Independently of execution policy, any pattern may opt into an optional `## Prompting Instructions` section (semantics: the `objective` skill's `references\u002Fprompting-instructions.md`). Offer it only when the interview surfaces durable prompting facts — standing context, validation gates, executor preference, standing hazards — rather than as a mandatory question.\n\n`roadmap.md` uses the umbrella skill's required headings, statuses, row sizing, validation-rows-as-evidence rule, and prose-guidance rules. Creation delta: when validation surfaces as a branch point in the interview, steer it into completion evidence under a semantic row — e.g. an indented `Evidence: targeted tests and relevant repo checks passed` — never a standalone final row.\n\n## Slug and path\n\n- Require an explicit slug, or propose a normalized slug and get explicit confirmation before writing any file.\n- Write only under `.ns\u002Fobjectives\u002F\u003Cslug>\u002F`; never `docs\u002Fobjectives\u002F` or anywhere else.\n- Slug identity is the umbrella skill's rule: renames never mint a new slug. Before creating a slug that looks like a rename or replacement of existing work, run `ns objective list --status all --format md`; if a likely existing Objective appears, stop per Stop \u002F ask.\n- Check the active root before writing. `ns objective exec read-objective \u003Cslug> --format md` returns a `not_found` envelope when the slug has no active record and otherwise emits it. If `.ns\u002Fobjectives\u002F\u003Cslug>\u002F` exists, stop per Stop \u002F ask — never overwrite. If the slug was previously deleted, source control history is the only historical link; recreate it only when the user wants that identity again.\n- Records are Markdown: write them directly, using `ns objective exec` only for deterministic reads; the only sanctioned YAML is Record Frontmatter below.\n\n## Record Frontmatter: initial edges and Blocked Sentence\n\nRecord Frontmatter mechanics — the exact-two-keys rule, mirrored two-file edits, and `ns objective check` — are owned by the `objective` umbrella skill. Creation deltas:\n\n- It is usually absent at creation: omit the block entirely unless the interview surfaces a real fact.\n- **Initial edges.** When the new Objective has a durable relationship to an existing record, declare the edge at creation — including the perspective-correct mirror entry in the counterpart's frontmatter, the one sanctioned edit outside the new record.\n- **Blocked Sentence.** Set `blocked:` only when the new record is genuinely gated at creation; the value is a non-empty sentence saying why.\n\n## Interview\n\nInterview the user relentlessly before writing (inspired by [Matt Pocock's `grill-me` skill](https:\u002F\u002Fgithub.com\u002Fmattpocock\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fproductivity\u002Fgrill-me\u002FSKILL.md)) until shared understanding covers title, thesis, scope, non-goals, completion criteria, assumptions, risks, open questions, and a semantic initial roadmap — or the user chooses to stop questioning and write.\n\n- **Pattern.** Early — once the thesis and work shape are roughly understood but before deep branch-walking — progressively disclose Objective shapes. Skip these menus when the user already named a pattern or the shape is unambiguous.\n  1. First present only two choices: **Create a standard Objective** (recommended; continue without a named pattern) and **Select an Objective template or pattern**. Do not name, summarize, or enumerate patterns in this first menu.\n  2. Only when the user selects the second choice, present a second numbered menu containing the six patterns, each with a one-line recognition cue sourced from the patterns catalog entries (do not restate deep prose), plus a final choice to go back and create a standard Objective.\n  3. Single-select the primary shaping; on pattern opt-in, read `references\u002F\u003Cpattern>-create.md` before asking that pattern's questions. Ask layering\u002Fcomposition questions only when the catalog says the chosen pattern composes.\n- Walk each branch of the design tree, resolving dependencies between decisions one by one; focus on branch points that affect scope, completion criteria, assumptions, risks, sequencing, closure evidence, or — only when the user explicitly requested execution behavior — execution policy.\n- Explore the codebase or existing docs instead of asking questions whose answers are discoverable locally.\n- Ask one unresolved question at a time, including your recommended answer so the user can confirm or correct it — framed so choosing it is an affirmative selection, never a \"no\" recommendation followed by \"Do you agree?\" — as a compact numbered menu with domain-specific labels — never an open-ended continuation prompt. Tell the user they can answer with a number or a custom correction.\n- Menu order: recommended path first, main alternative(s) next, and — only after the slug is explicitly confirmed — a final stop option reading exactly `Stop and create Objective \u003Cslug>`, confirmed slug verbatim. If the slug is unconfirmed, omit the stop option and resolve slug confirmation first.\n- Never write generic or invented durable content; keep interviewing until you don't have to.\n\n## Workflow\n\n1. Run the interview; confirm the slug and clear the active root per Slug and path, loading any conditional reference — a pattern's `references\u002F\u003Cpattern>-create.md` or `references\u002Fexecution-friendly-create.md` — before its questions.\n2. Create `.ns\u002Fobjectives\u002F\u003Cslug>\u002F` with `updates\u002F`, `objective.md`, and `roadmap.md` per Required shape, in concise human-readable narrative.\n3. If the interview surfaced initial edges or a genuine blocked gate, write Record Frontmatter per its section — including the counterpart mirrors — and run `ns objective check \u003Cslug>`; otherwise write no frontmatter.\n4. If the Objective is orienting — an agent doing unrelated work must obey its direction — write `orientation.md` (≈8 content lines, agent-facing) using the umbrella skill's format: `Direction`, `Getting to` (with ADR\u002FCONTEXT pointers), `What you see now`, `Avoid`, and `Active slice: see this objective's roadmap.md`. Otherwise skip it; presence of the file is the opt-in flag. Lifecycle\u002Fgraduation metadata stays in `roadmap.md`, never in `orientation.md`.\n5. Create no initial file under `updates\u002F` and no `closed.md`.\n\n## Stop \u002F ask\n\n- The slug is missing, unconfirmed, invalid-looking, or points outside `.ns\u002Fobjectives\u002F`.\n- The target Objective directory already exists — ask whether the user meant `objective-update` or a direct read.\n- The request looks like a rename\u002Freplacement of existing Objective work and the user has not explicitly chosen create vs update vs slug migration — ask whether they meant `objective-update`, a direct read, or an explicit slug migration.\n- Durable context is too thin to write thesis, scope, completion criteria, assumptions, risks, or requested execution policy without inventing them.\n- The request needs multiple Objectives: create only one and ask the user to run the command again for the others.\n\n## Verify\n\n- The directory contains `objective.md`, `roadmap.md`, and `updates\u002F`, and `objective.md` contains `## Assumptions and Risks`.\n- If orienting, `orientation.md` exists and follows the format; otherwise it is absent. It is optional, never required.\n- If execution-friendly, verify against `references\u002Fexecution-friendly-create.md`; if planning-only, execution policy sections are absent unless explicitly requested.\n- If a pattern was chosen, verify against that pattern reference's verification items.\n- If Record Frontmatter was written, it carries only `blocked` and\u002For `edges`, every declared edge has its mirror entry in the counterpart record, and `ns objective check \u003Cslug>` passes; if not written, `objective.md` starts with `# \u003CTitle>` and no frontmatter fence.\n- No files outside the new Objective's directory changed, except counterpart `objective.md` frontmatter blocks touched by mirrored edge entries.\n- There is no initial file under `updates\u002F` and no `closed.md`.\n- Summarize the created slug, first planned roadmap item, most important assumption or risk captured, and whether the Objective is planning-only or execution-friendly.\n","# objective-create\n\nCreate exactly one new Objective record under `.ns\u002Fobjectives\u002F\u003Cslug>\u002F`. Use the `objective` umbrella skill first for shared vocabulary, selection rules, storage model, safety boundaries, and family policy.\n\nThe Objective pattern set — wayfinding\u002Fideation, steelthread, standing, umbrella, autoobjective, and readme-driven-development — lives in `references\u002F\u003Cpattern>-create.md`; each reference owns that pattern's creation procedure on top of this skill's record mechanics. The pattern is chosen in the interview (see the Pattern beat below). When the user names a pattern upfront, read the matching reference immediately and skip the menu. Recognition prose and composition facts stay in the `objective` skill's patterns catalog (`references\u002Fobjective-patterns.md`).\n\n## Required shape\n\nNew records live in the active root only, `.ns\u002Fobjectives\u002F\u003Cslug>\u002F`, with `objective.md`, `roadmap.md`, and `updates\u002F` per the umbrella skill's storage model. Creation deltas: `orientation.md` is optional and orienting-only; there is no `closed.md` at creation.\n\n`objective.md` uses the umbrella skill's required headings.\n\n- `## Assumptions and Risks` must distinguish assumptions from risks, with enough context for a future `objective-update` to mark an assumption incorrect, a risk de-risked or not, or add newly discovered ones.\n- When the record accepts a deliberate shortcut — a prototype-race sacrifice of security, scope, or quality — never record the shortcut alone: name its upgrade beside it, so the later swap is a slice rather than a redesign and a future `objective-update` can retire the pair together.\n- Default to planning-only: execution-policy sections are never mandatory and appear only when the user explicitly asks for execution-friendly\u002Frunner\u002Fautonomous behavior. Never offer execution policy as an interview question on your own initiative — the Pattern beat settles this axis, and choosing a standard Objective (or any pattern other than autoobjective) without an explicit execution request means planning-only; execution policy can always be added later via `objective-update`. When the user does ask, read `references\u002Fexecution-friendly-create.md` before asking policy questions.\n- Independently of execution policy, any pattern may opt into an optional `## Prompting Instructions` section (semantics: the `objective` skill's `references\u002Fprompting-instructions.md`). Offer it only when the interview surfaces durable prompting facts — standing context, validation gates, executor preference, standing hazards — rather than as a mandatory question.\n\n`roadmap.md` uses the umbrella skill's required headings, statuses, row sizing, validation-rows-as-evidence rule, and prose-guidance rules. Creation delta: when validation surfaces as a branch point in the interview, steer it into completion evidence under a semantic row — e.g. an indented `Evidence: targeted tests and relevant repo checks passed` — never a standalone final row.\n\n## Slug and path\n\n- Require an explicit slug, or propose a normalized slug and get explicit confirmation before writing any file.\n- Write only under `.ns\u002Fobjectives\u002F\u003Cslug>\u002F`; never `docs\u002Fobjectives\u002F` or anywhere else.\n- Slug identity is the umbrella skill's rule: renames never mint a new slug. Before creating a slug that looks like a rename or replacement of existing work, run `ns objective list --status all --format md`; if a likely existing Objective appears, stop per Stop \u002F ask.\n- Check the active root before writing. `ns objective exec read-objective \u003Cslug> --format md` returns a `not_found` envelope when the slug has no active record and otherwise emits it. If `.ns\u002Fobjectives\u002F\u003Cslug>\u002F` exists, stop per Stop \u002F ask — never overwrite. If the slug was previously deleted, source control history is the only historical link; recreate it only when the user wants that identity again.\n- Records are Markdown: write them directly, using `ns objective exec` only for deterministic reads; the only sanctioned YAML is Record Frontmatter below.\n\n## Record Frontmatter: initial edges and Blocked Sentence\n\nRecord Frontmatter mechanics — the exact-two-keys rule, mirrored two-file edits, and `ns objective check` — are owned by the `objective` umbrella skill. Creation deltas:\n\n- It is usually absent at creation: omit the block entirely unless the interview surfaces a real fact.\n- **Initial edges.** When the new Objective has a durable relationship to an existing record, declare the edge at creation — including the perspective-correct mirror entry in the counterpart's frontmatter, the one sanctioned edit outside the new record.\n- **Blocked Sentence.** Set `blocked:` only when the new record is genuinely gated at creation; the value is a non-empty sentence saying why.\n\n## Interview\n\nInterview the user relentlessly before writing (inspired by [Matt Pocock's `grill-me` skill](https:\u002F\u002Fgithub.com\u002Fmattpocock\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fproductivity\u002Fgrill-me\u002FSKILL.md)) until shared understanding covers title, thesis, scope, non-goals, completion criteria, assumptions, risks, open questions, and a semantic initial roadmap — or the user chooses to stop questioning and write.\n\n- **Pattern.** Early — once the thesis and work shape are roughly understood but before deep branch-walking — progressively disclose Objective shapes. Skip these menus when the user already named a pattern or the shape is unambiguous.\n  1. First present only two choices: **Create a standard Objective** (recommended; continue without a named pattern) and **Select an Objective template or pattern**. Do not name, summarize, or enumerate patterns in this first menu.\n  2. Only when the user selects the second choice, present a second numbered menu containing the six patterns, each with a one-line recognition cue sourced from the patterns catalog entries (do not restate deep prose), plus a final choice to go back and create a standard Objective.\n  3. Single-select the primary shaping; on pattern opt-in, read `references\u002F\u003Cpattern>-create.md` before asking that pattern's questions. Ask layering\u002Fcomposition questions only when the catalog says the chosen pattern composes.\n- Walk each branch of the design tree, resolving dependencies between decisions one by one; focus on branch points that affect scope, completion criteria, assumptions, risks, sequencing, closure evidence, or — only when the user explicitly requested execution behavior — execution policy.\n- Explore the codebase or existing docs instead of asking questions whose answers are discoverable locally.\n- Ask one unresolved question at a time, including your recommended answer so the user can confirm or correct it — framed so choosing it is an affirmative selection, never a \"no\" recommendation followed by \"Do you agree?\" — as a compact numbered menu with domain-specific labels — never an open-ended continuation prompt. Tell the user they can answer with a number or a custom correction.\n- Menu order: recommended path first, main alternative(s) next, and — only after the slug is explicitly confirmed — a final stop option reading exactly `Stop and create Objective \u003Cslug>`, confirmed slug verbatim. If the slug is unconfirmed, omit the stop option and resolve slug confirmation first.\n- Never write generic or invented durable content; keep interviewing until you don't have to.\n\n## Workflow\n\n1. Run the interview; confirm the slug and clear the active root per Slug and path, loading any conditional reference — a pattern's `references\u002F\u003Cpattern>-create.md` or `references\u002Fexecution-friendly-create.md` — before its questions.\n2. Create `.ns\u002Fobjectives\u002F\u003Cslug>\u002F` with `updates\u002F`, `objective.md`, and `roadmap.md` per Required shape, in concise human-readable narrative.\n3. If the interview surfaced initial edges or a genuine blocked gate, write Record Frontmatter per its section — including the counterpart mirrors — and run `ns objective check \u003Cslug>`; otherwise write no frontmatter.\n4. If the Objective is orienting — an agent doing unrelated work must obey its direction — write `orientation.md` (≈8 content lines, agent-facing) using the umbrella skill's format: `Direction`, `Getting to` (with ADR\u002FCONTEXT pointers), `What you see now`, `Avoid`, and `Active slice: see this objective's roadmap.md`. Otherwise skip it; presence of the file is the opt-in flag. Lifecycle\u002Fgraduation metadata stays in `roadmap.md`, never in `orientation.md`.\n5. Create no initial file under `updates\u002F` and no `closed.md`.\n\n## Stop \u002F ask\n\n- The slug is missing, unconfirmed, invalid-looking, or points outside `.ns\u002Fobjectives\u002F`.\n- The target Objective directory already exists — ask whether the user meant `objective-update` or a direct read.\n- The request looks like a rename\u002Freplacement of existing Objective work and the user has not explicitly chosen create vs update vs slug migration — ask whether they meant `objective-update`, a direct read, or an explicit slug migration.\n- Durable context is too thin to write thesis, scope, completion criteria, assumptions, risks, or requested execution policy without inventing them.\n- The request needs multiple Objectives: create only one and ask the user to run the command again for the others.\n\n## Verify\n\n- The directory contains `objective.md`, `roadmap.md`, and `updates\u002F`, and `objective.md` contains `## Assumptions and Risks`.\n- If orienting, `orientation.md` exists and follows the format; otherwise it is absent. It is optional, never required.\n- If execution-friendly, verify against `references\u002Fexecution-friendly-create.md`; if planning-only, execution policy sections are absent unless explicitly requested.\n- If a pattern was chosen, verify against that pattern reference's verification items.\n- If Record Frontmatter was written, it carries only `blocked` and\u002For `edges`, every declared edge has its mirror entry in the counterpart record, and `ns objective check \u003Cslug>` passes; if not written, `objective.md` starts with `# \u003CTitle>` and no frontmatter fence.\n- No files outside the new Objective's directory changed, except counterpart `objective.md` frontmatter blocks touched by mirrored edge entries.\n- There is no initial file under `updates\u002F` and no `closed.md`.\n- Summarize the created slug, first planned roadmap item, most important assumption or risk captured, and whether the Objective is planning-only or execution-friendly.\n",{"name":5,"description":7,"disable-model-invocation":44},true,"master","44a56f0d543f274b42239091b35fe4d23855f6b3","skills\u002Fincubating\u002Fobjectives\u002Fobjective-create",0,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns","https:\u002F\u002Fgithub.com\u002Fnseng-ai","Keeping the clankers in check: Nonslop Engineering for AI","2026-10-02 12:48:53",{"source":8,"skillName":5,"commands":54,"prompt":64,"zipAvailable":44},[55,58,61],{"tool":56,"command":57},"npx","npx skills add nseng-ai\u002Fns --skill objective-create",{"tool":59,"command":60},"bunx","bunx skills add nseng-ai\u002Fns --skill objective-create",{"tool":62,"command":63},"pnpm","pnpm dlx skills add nseng-ai\u002Fns --skill objective-create","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002FSKILL.md",[66,72,76,80,84,88,92,96],{"path":67,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":70,"sourceHtmlUrl":71},"agents\u002Fopenai.yaml","REFERENCE","参考文档",43,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Fagents\u002Fopenai.yaml",{"path":73,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fautoobjective-create.md",2506,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Fautoobjective-create.md",{"path":77,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fexecution-friendly-create.md",2272,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Fexecution-friendly-create.md",{"path":81,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Freadme-driven-development-create.md",2730,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Freadme-driven-development-create.md",{"path":85,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Fstanding-create.md",1314,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Fstanding-create.md",{"path":89,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002Fsteelthread-create.md",2556,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Fsteelthread-create.md",{"path":93,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Fumbrella-create.md",2186,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Fumbrella-create.md",{"path":97,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":98,"sourceHtmlUrl":99},"references\u002Fwayfinding-create.md",2992,"https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fincubating\u002Fobjectives\u002Fobjective-create\u002Freferences\u002Fwayfinding-create.md",[101,123,153,176,199,210],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":39},"01M6KM3VZ7JC4S6R4KATJ368EH","pr-make-accountable","Interview a PR author to establish shared understanding of intent and design decisions, then co-author the What \u002F Why \u002F Changes PR body with opt-in Requested reviewer focus.","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Fpublic\u002Fprs\u002Fpr-make-accountable\u002FSKILL.md",[],[108,109,110,111,112,113,114,115,116,117,118,119,120,24,121,122],"pr","make","pr make accountable","design","why","reviewer","decisions","interview","intent","focus","shared","co","with","changes","pr body",{"skillNo":124,"name":125,"nameZh":6,"description":126,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":134,"publishedAt":39},"01M6KM3NZCZ86VJ91RSVTRGTJW","typescript-style","Opinionated style guide for strict, portable TypeScript. Use when writing, designing, or reviewing TypeScript: type design, boundary validation, error handling, module architecture, and review discipline.","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Finternal\u002Ftypescript\u002Ftypescript-style\u002FSKILL.md",[129],{"categoryNo":130,"categoryName":131,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",[135,136,137,111,138,139,140,141,142,143,144,145,36,146,147,148,149,150,151,152],"typescript","style","typescript style","writing","review","architecture","core","error handling","system","Rules","README","error-handling","study","module","error","case-study","style guide","for",{"skillNo":154,"name":155,"nameZh":6,"description":156,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":157,"categories":158,"keywords":162,"publishedAt":39},"01M6KM3HNY4WY42KMYRPS6KEJZ","typescript-fake-driven-testing","TypeScript fake-driven testing architecture. Use when writing or reviewing TS code with external dependencies, gateway interfaces, real adapters, in-memory fakes, or CLI scenario tests.","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Finternal\u002Ftypescript\u002Ftypescript-fake-driven-testing\u002FSKILL.md",[159],{"categoryNo":160,"categoryName":161,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000008","测试与质量",[163,135,164,165,166,138,167,168,140,169,170,36,171,172,173,174,120,175],"testing","driven","fake driven testing","typescript fake driven testing","cli","memory","code","gateway","dependencies","tests","scenario","Adapters","in-memory",{"skillNo":177,"name":178,"nameZh":6,"description":179,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":180,"categories":181,"keywords":182,"publishedAt":39},"01M6KM3AXC2539PENYVZJYKCP9","ns-typescript","TypeScript overlay for ns-style projects. Use when editing or reviewing TS in this repo: tsconfig baseline, pnpm\u002FVitest\u002Foxlint\u002Foxfmt\u002Fnative tsc commands, relative .ts imports, workspace subpath exports, and the exactOptionalPropertyTypes spread idiom. Repo-specific enforcement (test lanes, time seams, style guard) lives in the host repo's AGENTS.md.","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Finternal\u002Ftypescript\u002Fns-typescript\u002FSKILL.md",[],[135,183,184,185,186,187,188,189,36,190,191,192,193,136,194,195,196,152,197,198],"ns typescript","test","guard","vitest","projects","editing","repo","enforcement","references","time","host","oxlint","import","native","alternatives","imports",{"skillNo":200,"name":201,"nameZh":6,"description":202,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":203,"categories":204,"keywords":205,"publishedAt":39},"01M6KM36Y5KZ5V48M58MT42AHM","ns-typescript-style-tripwire","Invoke the ns TypeScript style Tripwire against a supplied diff or current branch.","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Finternal\u002Ftypescript\u002Fns-typescript-style-tripwire\u002FSKILL.md",[],[135,206,136,207,208,209,24],"tripwire","ns typescript style tripwire","branch","diff",{"skillNo":211,"name":212,"nameZh":6,"description":213,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":214,"categories":215,"keywords":219,"publishedAt":39},"01M6KM313HANS0AV181X37ZM45","skill-management","Manage skills with `npx skills`: add, edit, remove, rename, update, or list skills (local or GitHub), and the nested `skills\u002F\u003Cdisposition>\u002F...\u002F\u003Cname>\u002F` \u002F flat `.agents\u002Fskills\u002F` layout conventions.","https:\u002F\u002Fgithub.com\u002Fnseng-ai\u002Fns\u002Fblob\u002Fmaster\u002Fskills\u002Finternal\u002Fskill-system\u002Fskill-management\u002FSKILL.md",[216],{"categoryNo":217,"categoryName":218,"taxonomyType":132,"taxonomyTypeDesc":133},"01JBSEEDCAT000000000000023","Skill与智能体开发",[220,221,222,223,224,209,225,226,56,227,228,120,229,230,231,232,233,234,24,235],"management","skill management","github","Bash","remove","nested","manage","local","cat","ln","ls","mkdir","mv","readlink","rg","lock"]