[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75Y2JCJ83TXD4A2Z37CG97P":3,"skill-same:01M75Y2JCJ83TXD4A2Z37CG97P":59},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"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":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":44},"01M75Y2JCJ83TXD4A2Z37CG97P","feature",null,"Turn the next, named, or numbered build-plan feature into a buildable current-feature.md spec with small steps and done-when criteria. Use for \u002Ffeature, planning a feature, or starting planned work.","aiblueprinthq\u002Fai-blueprint","aiblueprinthq","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F317905365?v=4","01M75XZGKSB8GPVXQ69BA57VA7","01M75XZGME4HNZG9CDB8K3TQ8X",510,"MIT","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffeature\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: feature\ndescription: Turn the next, named, or numbered build-plan feature into a buildable current-feature.md spec with small steps and done-when criteria. Use for \u002Ffeature, planning a feature, or starting planned work.\ndisable-model-invocation: true\n---\n\n# feature - create the active implementation spec\n\n**Context reuse:** Reuse any required file already loaded in project instructions or the current session. Read it again only if absent, changed, or exact current bytes or line references are needed.\n\nThis skill plans one feature and stops before implementation. Its output is\n`blueprint\u002Fcontext\u002Fcurrent-feature.md`.\n\n## Start\n\n**First action:** Before project inspection, preflight, or any other tool call,\npublish the `feature` activity as `running` when `blueprint\u002F.state\u002F` exists.\nFrom the project root, run the first helper that exists,\n`node .agents\u002Fskills\u002Fblueprint-doctor\u002Fscripts\u002Frun-state.mjs` or\n`node .claude\u002Fskills\u002Fblueprint-doctor\u002Fscripts\u002Frun-state.mjs`, with `start --command\n\u003Cskill> --summary \"\u003Ctruthful summary>\" --boundary\n\u003Cread-only|reviewed|local-only>`. Use `update` at milestones, with `--status\nblocked --resume \u003Ccommand>` for a blocker, and `finish --status ready|completed`\nat the end; `--help` lists every option and `blueprint\u002Fworkflow.md` has the full\nactivity contract. A missing or failing helper is a warning, not a blocker.\nCombine that write with the first context-gathering tool batch when the adapter\nsupports it.\n\nRead `blueprint\u002Fconfig.json` only for settings that affect the spec. Invalid\nconfiguration stops mutating work and points to `\u002Fblueprint-doctor`.\n\nConfirm that `blueprint\u002Fcontext\u002Fcurrent-feature.md` is the empty stub. If it\ncontains active work, stop and direct the user to resume or complete it. Never\nreplace another worker's active spec. Parallel work uses one clone or Git\nworktree per work item, with a separate branch in each checkout.\n\nResolve the target from `blueprint\u002Fbuild-plan.md`:\n\n- Use the requested number or name when it matches a checklist item.\n- With no argument, use the first unchecked leaf item.\n- Read only the matching checklist line, its parent, and nearby text needed to\n  understand the hierarchy.\n- If a plain list has no checkboxes, treat its first item as unchecked and offer\n  to convert the plan after the spec review.\n- If the requested item is already checked, stop. It was built. To build it\n  again, the user first removes it through `\u002Ffix` and unchecks the item.\n\nA `--check` argument requests the spec check for this run and is not part of\nthe target. State the selected feature in one sentence.\n\n## Build one authoritative feature packet\n\nGather the smallest packet that can answer what must be built:\n\n1. Search `blueprint\u002Fcontext\u002Fproject-overview.md` for the feature number, title,\n   and distinctive nouns from the target line. Read the matching feature\n   passage plus only the usage-model, data-model, stack, UI, security, or\n   deployment passages it directly depends on. Do not read the whole overview\n   by default.\n2. Inspect the repository once, starting from paths named by those passages.\n   Follow only relevant imports, callers, tests, schemas, and configuration.\n   Batch related searches and reads when supported.\n3. Use the Commands section already loaded from `AGENTS.md`. Read it from disk\n   only when it is absent or changed. Read only applicable sections of\n   `blueprint\u002Fcontext\u002Fcoding-standards.md`.\n4. Run the declared Verify command once when it exists. Record only observed\n   results. Do not start a dev server.\n\nFinish context gathering in at most four tool rounds after this skill starts:\ntarget and overview matches, one batched repository inspection, applicable\nstandards only if needed, and Verify. Combine or skip rounds when possible. Do\nnot inspect other skill directories, `ai-interaction.md`, findings, review records,\nor templates during normal planned-feature work. The only history exception is\nthis skill's `reference\u002Fbuild-history.md` and the selected feature's archive metadata\nneeded to freeze its build attempt below; batch this with the target lookup,\nwithout loading unrelated history. Do not create\nscratch code or run implementation probes while writing a spec. Put a check in the\nrelevant build step when a repository detail cannot be confirmed from existing\nevidence.\n\nThe plans and overview define product intent. The repository defines current\nreality. Do not invent presets, defaults, limits, permissions, money rules,\ndestructive behavior, stored fields, or API contracts. A familiar label is not a\ncomplete contract when it has multiple reasonable meanings. Put unresolved\nmaterial choices under an `Open questions` heading and in the review handoff.\nStop without writing the spec when implementation cannot begin safely until one\nis answered. Do not block on a reversible internal implementation detail with no\nuser-visible, security, persisted-data, or interoperability consequence. Choose\nthe simplest repository-native option, record it in the spec, and require a test\nseam when the value is nondeterministic. Planned future persistence alone does\nnot make a current in-memory representation a product decision when no stored\ndata or external compatibility exists yet.\n\nApply proportional engineering before drafting: add an abstraction, dependency,\nservice, configuration surface, compatibility layer, or security mechanism only\nwhen an established requirement needs it now. Prefer existing code, the standard\nlibrary, native platform features, and installed dependencies. Unknown scale or\nfuture extensibility defaults to the smaller reversible design. Treat a trust or\ndata-integrity boundary as established when the repository exposes network or\nuntrusted input, auth\u002Fsession\u002Fownership, shared persisted data, destructive\noperations, secrets, or sensitive data, even when the plans do not name it.\n\nIf `project-overview.md` is 20,000 bytes or larger, stop and ask for `\u002Foverview`\ninstead of loading it. If the target is too large for one reviewable branch,\npropose sub-features and wait for approval before editing the user-owned build\nplan. After approval, add lettered checklist items under the parent and spec only\nthe first one.\n\n## New feature not in the plan\n\nDo not silently add scope. Search for a duplicate, then propose one checkbox line\nand its placement. Include a `project-plan.md` edit only when the request changes\nthe product direction, users, data, stack, monetization, UI, or deployment. Wait\nfor approval, update the plans, run the installed `overview` skill, and then\nresume this skill. Bugs and small unplanned changes belong in `\u002Ffix`.\n\n## Write the final spec once\n\nBefore review, allocate the selected stable feature ID's build attempt using\n`reference\u002Fbuild-history.md`: first build 1, otherwise one greater than\nthe highest attempt archived for that ID. Preserve the ID across renamed titles\nand lettered sub-items. Stop on ambiguous history; never infer attempts from a\ntitle suffix, file count, or timestamps.\n\nDraft and critique in context, then write\n`blueprint\u002Fcontext\u002Fcurrent-feature.md` once. A later write is only for a\nmechanical correction, a selected spec check, or a user-requested revision. Record `**Branch:**` with the\nfull configured feature branch. The first heading and build-plan identity must\nuse this canonical form:\n\n```markdown\n# Feature: \u003Ctitle>\n\n**From build-plan:** feature \u003Cid>\n**Build attempt:** \u003Cpositive integer>\n```\n\nThen use these section headings:\n\n- Goal\n- Design reference, only for visual or replication work\n- In scope\n- Out of scope\n- Build loop\n- Build steps\n- Files \u002F areas\n- Data \u002F contracts\n- Testing\n- Notes for the AI\n- Open questions, only when a product decision remains unresolved\n\nBuild steps are ordered checklist items. Each step must leave the project\nworking, stay small enough to review, and end with a concrete `Done when` that\nnames observable behavior and the relevant check. Follow `workflow.stepReview`\nand `workflow.checkpointCommits` from config in the Build loop, and describe a\ncheckpoint commit as a per-step choice the user picks in `\u002Fimplement`, never an\nautomatic step. `\u002Fcomplete` creates the final feature commit.\n\nThe spec must preserve every explicit contract in the feature packet, including\napplicable project-wide UX and security requirements. Do not discard a required\nstate because the current fixture cannot trigger it yet. Keep later features out,\ndefine authorization and tenant boundaries only when the feature packet or\nreachable code establishes them, identify client and server responsibilities,\nand name exact files or areas supported by repository evidence.\nAdd focused tests for logic when a test command exists. Add browser coverage only\nwhen a Browser tests command exists and it is proportionate. Do not claim live,\nvisual, persisted-data, or integration evidence that was not run.\n\nBuild the branch value from the configured feature prefix plus the feature title\nin lowercase kebab-case. Replace each run of characters other than ASCII letters\nand digits with one hyphen and trim edge hyphens. For attempt N > 1, append\n`--build-N` to that slug before recording the full branch, for example\n`feature\u002Fexport-reports--build-2`. The reserved double hyphen distinguishes the\nattempt from a title ending in `Build 2`. Before freezing the new spec, check the\nexact archive path and branch availability using `reference\u002Fbuild-history.md`.\nStop on filesystem entries or disallowed refs; never auto-bump the attempt. The reference permits only one existing-ref\ncase: the pristine current branch of a pre-created parallel worktree. Freeze both\nfields before review; completion and resume reuse them rather than allocating\nagain.\n\nFor visual replication, require an existing screenshot or reference. Store a\nprovided image under `blueprint\u002Freference\u002F` and link it. If `prototypes\u002F` exists,\nuse its relevant HTML and `theme.css`; port shared tokens before feature UI.\n\n## Critique gate\n\nBefore the single write, check these failure classes:\n\n- Missing happy, loading, empty, invalid, denied, and unexpected-error behavior\n  that applies to the feature.\n- A product contract from the packet that was omitted, weakened, or contradicted.\n- Scope added from guesswork or pulled forward from a later feature.\n- A proposed abstraction, dependency, service, configuration surface,\n  compatibility layer, or security mechanism lacks a current requirement, or\n  duplicates existing code, the standard library, the platform, or an installed\n  dependency. Untuned stack-specific standards in `coding-standards.md` are not\n  established requirements.\n- An oversized or incorrectly ordered build step.\n- When an established persisted-data or external API boundary requires it, the\n  contract leaves a material type, format, encoding, generator, uniqueness rule,\n  default, lifecycle, serialization, or stable result and error shape implicit.\n- When an established security, tenant, concurrency, destructive-operation,\n  payment, or sensitive-data boundary requires it, the trusted actor source,\n  repository-first tenant scope, atomicity, idempotency, or redaction behavior\n  remains implicit.\n- User-controlled text lacks a safe rendering rule, or validation and error\n  feedback lacks the relevant label, association, announcement, focus, or\n  clearing behavior.\n- A new file, asset, import, or route is not reachable through the current\n  runtime and server behavior.\n- An authorization or URL shape later work would have to reinterpret.\n- A done-when that cannot be observed or a test claim the repository cannot run.\n- A prerequisite that makes the final Verify gate impossible.\n\nIf a prerequisite is absent, incomplete, untracked, or unverified, say which one\nthe evidence shows. Do not bury repair inside this feature. Stop with the exact\n`\u002Ffix` or user decision required.\n\nOtherwise write the tightened spec. When `qualityGates.regular.specCheck` is\n`always` or this run requested `--check`, follow `reference\u002Fspec-check.md`\nbefore the review stop. Then update activity to `ready` and stop for review.\nLead with a short note naming what the critique and any spec check changed, or\nsay that they found no material change. Never implement from this skill.\n","# feature - create the active implementation spec\n\n**Context reuse:** Reuse any required file already loaded in project instructions or the current session. Read it again only if absent, changed, or exact current bytes or line references are needed.\n\nThis skill plans one feature and stops before implementation. Its output is\n`blueprint\u002Fcontext\u002Fcurrent-feature.md`.\n\n## Start\n\n**First action:** Before project inspection, preflight, or any other tool call,\npublish the `feature` activity as `running` when `blueprint\u002F.state\u002F` exists.\nFrom the project root, run the first helper that exists,\n`node .agents\u002Fskills\u002Fblueprint-doctor\u002Fscripts\u002Frun-state.mjs` or\n`node .claude\u002Fskills\u002Fblueprint-doctor\u002Fscripts\u002Frun-state.mjs`, with `start --command\n\u003Cskill> --summary \"\u003Ctruthful summary>\" --boundary\n\u003Cread-only|reviewed|local-only>`. Use `update` at milestones, with `--status\nblocked --resume \u003Ccommand>` for a blocker, and `finish --status ready|completed`\nat the end; `--help` lists every option and `blueprint\u002Fworkflow.md` has the full\nactivity contract. A missing or failing helper is a warning, not a blocker.\nCombine that write with the first context-gathering tool batch when the adapter\nsupports it.\n\nRead `blueprint\u002Fconfig.json` only for settings that affect the spec. Invalid\nconfiguration stops mutating work and points to `\u002Fblueprint-doctor`.\n\nConfirm that `blueprint\u002Fcontext\u002Fcurrent-feature.md` is the empty stub. If it\ncontains active work, stop and direct the user to resume or complete it. Never\nreplace another worker's active spec. Parallel work uses one clone or Git\nworktree per work item, with a separate branch in each checkout.\n\nResolve the target from `blueprint\u002Fbuild-plan.md`:\n\n- Use the requested number or name when it matches a checklist item.\n- With no argument, use the first unchecked leaf item.\n- Read only the matching checklist line, its parent, and nearby text needed to\n  understand the hierarchy.\n- If a plain list has no checkboxes, treat its first item as unchecked and offer\n  to convert the plan after the spec review.\n- If the requested item is already checked, stop. It was built. To build it\n  again, the user first removes it through `\u002Ffix` and unchecks the item.\n\nA `--check` argument requests the spec check for this run and is not part of\nthe target. State the selected feature in one sentence.\n\n## Build one authoritative feature packet\n\nGather the smallest packet that can answer what must be built:\n\n1. Search `blueprint\u002Fcontext\u002Fproject-overview.md` for the feature number, title,\n   and distinctive nouns from the target line. Read the matching feature\n   passage plus only the usage-model, data-model, stack, UI, security, or\n   deployment passages it directly depends on. Do not read the whole overview\n   by default.\n2. Inspect the repository once, starting from paths named by those passages.\n   Follow only relevant imports, callers, tests, schemas, and configuration.\n   Batch related searches and reads when supported.\n3. Use the Commands section already loaded from `AGENTS.md`. Read it from disk\n   only when it is absent or changed. Read only applicable sections of\n   `blueprint\u002Fcontext\u002Fcoding-standards.md`.\n4. Run the declared Verify command once when it exists. Record only observed\n   results. Do not start a dev server.\n\nFinish context gathering in at most four tool rounds after this skill starts:\ntarget and overview matches, one batched repository inspection, applicable\nstandards only if needed, and Verify. Combine or skip rounds when possible. Do\nnot inspect other skill directories, `ai-interaction.md`, findings, review records,\nor templates during normal planned-feature work. The only history exception is\nthis skill's `reference\u002Fbuild-history.md` and the selected feature's archive metadata\nneeded to freeze its build attempt below; batch this with the target lookup,\nwithout loading unrelated history. Do not create\nscratch code or run implementation probes while writing a spec. Put a check in the\nrelevant build step when a repository detail cannot be confirmed from existing\nevidence.\n\nThe plans and overview define product intent. The repository defines current\nreality. Do not invent presets, defaults, limits, permissions, money rules,\ndestructive behavior, stored fields, or API contracts. A familiar label is not a\ncomplete contract when it has multiple reasonable meanings. Put unresolved\nmaterial choices under an `Open questions` heading and in the review handoff.\nStop without writing the spec when implementation cannot begin safely until one\nis answered. Do not block on a reversible internal implementation detail with no\nuser-visible, security, persisted-data, or interoperability consequence. Choose\nthe simplest repository-native option, record it in the spec, and require a test\nseam when the value is nondeterministic. Planned future persistence alone does\nnot make a current in-memory representation a product decision when no stored\ndata or external compatibility exists yet.\n\nApply proportional engineering before drafting: add an abstraction, dependency,\nservice, configuration surface, compatibility layer, or security mechanism only\nwhen an established requirement needs it now. Prefer existing code, the standard\nlibrary, native platform features, and installed dependencies. Unknown scale or\nfuture extensibility defaults to the smaller reversible design. Treat a trust or\ndata-integrity boundary as established when the repository exposes network or\nuntrusted input, auth\u002Fsession\u002Fownership, shared persisted data, destructive\noperations, secrets, or sensitive data, even when the plans do not name it.\n\nIf `project-overview.md` is 20,000 bytes or larger, stop and ask for `\u002Foverview`\ninstead of loading it. If the target is too large for one reviewable branch,\npropose sub-features and wait for approval before editing the user-owned build\nplan. After approval, add lettered checklist items under the parent and spec only\nthe first one.\n\n## New feature not in the plan\n\nDo not silently add scope. Search for a duplicate, then propose one checkbox line\nand its placement. Include a `project-plan.md` edit only when the request changes\nthe product direction, users, data, stack, monetization, UI, or deployment. Wait\nfor approval, update the plans, run the installed `overview` skill, and then\nresume this skill. Bugs and small unplanned changes belong in `\u002Ffix`.\n\n## Write the final spec once\n\nBefore review, allocate the selected stable feature ID's build attempt using\n`reference\u002Fbuild-history.md`: first build 1, otherwise one greater than\nthe highest attempt archived for that ID. Preserve the ID across renamed titles\nand lettered sub-items. Stop on ambiguous history; never infer attempts from a\ntitle suffix, file count, or timestamps.\n\nDraft and critique in context, then write\n`blueprint\u002Fcontext\u002Fcurrent-feature.md` once. A later write is only for a\nmechanical correction, a selected spec check, or a user-requested revision. Record `**Branch:**` with the\nfull configured feature branch. The first heading and build-plan identity must\nuse this canonical form:\n\n```markdown\n# Feature: \u003Ctitle>\n\n**From build-plan:** feature \u003Cid>\n**Build attempt:** \u003Cpositive integer>\n```\n\nThen use these section headings:\n\n- Goal\n- Design reference, only for visual or replication work\n- In scope\n- Out of scope\n- Build loop\n- Build steps\n- Files \u002F areas\n- Data \u002F contracts\n- Testing\n- Notes for the AI\n- Open questions, only when a product decision remains unresolved\n\nBuild steps are ordered checklist items. Each step must leave the project\nworking, stay small enough to review, and end with a concrete `Done when` that\nnames observable behavior and the relevant check. Follow `workflow.stepReview`\nand `workflow.checkpointCommits` from config in the Build loop, and describe a\ncheckpoint commit as a per-step choice the user picks in `\u002Fimplement`, never an\nautomatic step. `\u002Fcomplete` creates the final feature commit.\n\nThe spec must preserve every explicit contract in the feature packet, including\napplicable project-wide UX and security requirements. Do not discard a required\nstate because the current fixture cannot trigger it yet. Keep later features out,\ndefine authorization and tenant boundaries only when the feature packet or\nreachable code establishes them, identify client and server responsibilities,\nand name exact files or areas supported by repository evidence.\nAdd focused tests for logic when a test command exists. Add browser coverage only\nwhen a Browser tests command exists and it is proportionate. Do not claim live,\nvisual, persisted-data, or integration evidence that was not run.\n\nBuild the branch value from the configured feature prefix plus the feature title\nin lowercase kebab-case. Replace each run of characters other than ASCII letters\nand digits with one hyphen and trim edge hyphens. For attempt N > 1, append\n`--build-N` to that slug before recording the full branch, for example\n`feature\u002Fexport-reports--build-2`. The reserved double hyphen distinguishes the\nattempt from a title ending in `Build 2`. Before freezing the new spec, check the\nexact archive path and branch availability using `reference\u002Fbuild-history.md`.\nStop on filesystem entries or disallowed refs; never auto-bump the attempt. The reference permits only one existing-ref\ncase: the pristine current branch of a pre-created parallel worktree. Freeze both\nfields before review; completion and resume reuse them rather than allocating\nagain.\n\nFor visual replication, require an existing screenshot or reference. Store a\nprovided image under `blueprint\u002Freference\u002F` and link it. If `prototypes\u002F` exists,\nuse its relevant HTML and `theme.css`; port shared tokens before feature UI.\n\n## Critique gate\n\nBefore the single write, check these failure classes:\n\n- Missing happy, loading, empty, invalid, denied, and unexpected-error behavior\n  that applies to the feature.\n- A product contract from the packet that was omitted, weakened, or contradicted.\n- Scope added from guesswork or pulled forward from a later feature.\n- A proposed abstraction, dependency, service, configuration surface,\n  compatibility layer, or security mechanism lacks a current requirement, or\n  duplicates existing code, the standard library, the platform, or an installed\n  dependency. Untuned stack-specific standards in `coding-standards.md` are not\n  established requirements.\n- An oversized or incorrectly ordered build step.\n- When an established persisted-data or external API boundary requires it, the\n  contract leaves a material type, format, encoding, generator, uniqueness rule,\n  default, lifecycle, serialization, or stable result and error shape implicit.\n- When an established security, tenant, concurrency, destructive-operation,\n  payment, or sensitive-data boundary requires it, the trusted actor source,\n  repository-first tenant scope, atomicity, idempotency, or redaction behavior\n  remains implicit.\n- User-controlled text lacks a safe rendering rule, or validation and error\n  feedback lacks the relevant label, association, announcement, focus, or\n  clearing behavior.\n- A new file, asset, import, or route is not reachable through the current\n  runtime and server behavior.\n- An authorization or URL shape later work would have to reinterpret.\n- A done-when that cannot be observed or a test claim the repository cannot run.\n- A prerequisite that makes the final Verify gate impossible.\n\nIf a prerequisite is absent, incomplete, untracked, or unverified, say which one\nthe evidence shows. Do not bury repair inside this feature. Stop with the exact\n`\u002Ffix` or user decision required.\n\nOtherwise write the tightened spec. When `qualityGates.regular.specCheck` is\n`always` or this run requested `--check`, follow `reference\u002Fspec-check.md`\nbefore the review stop. Then update activity to `ready` and stop for review.\nLead with a short note naming what the critique and any spec check changed, or\nsay that they found no material change. Never implement from this skill.\n",{"name":5,"description":7,"disable-model-invocation":23},true,"main","2ac6d59a4c57a9d69ce652adca1976d8ef8f182c",".claude\u002Fskills\u002Ffeature",132,"https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint","https:\u002F\u002Fai-blueprint.dev","https:\u002F\u002Fgithub.com\u002Faiblueprinthq","2026-11-14 19:46:05",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":23},[34,37,40],{"tool":35,"command":36},"npx","npx skills add aiblueprinthq\u002Fai-blueprint --skill feature",{"tool":38,"command":39},"bunx","bunx skills add aiblueprinthq\u002Fai-blueprint --skill feature",{"tool":41,"command":42},"pnpm","pnpm dlx skills add aiblueprinthq\u002Fai-blueprint --skill feature","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffeature\u002FSKILL.md",[45,51,55],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"reference\u002Fbuild-history.md","REFERENCE","参考文档",2413,"https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffeature\u002Freference\u002Fbuild-history.md",{"path":52,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":53,"sourceHtmlUrl":54},"reference\u002Ffeature-spec-template.md",2819,"https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffeature\u002Freference\u002Ffeature-spec-template.md",{"path":56,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":57,"sourceHtmlUrl":58},"reference\u002Fspec-check.md",3671,"https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffeature\u002Freference\u002Fspec-check.md",[60,66,71,78,85,92],{"skillNo":61,"name":62,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":18},"01M75Y34X26C74A66RYE49W8JY","status","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002Fpackages\u002Fcreate-ai-blueprint\u002Ftest\u002Ffixtures\u002Fhealthy-blueprint\u002F.claude\u002Fskills\u002Fstatus\u002FSKILL.md",[],[],{"skillNo":67,"name":62,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":18},"01M75Y3456H4WZZ25634QY9KRA","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002Fpackages\u002Fcreate-ai-blueprint\u002Ftest\u002Ffixtures\u002Fhealthy-blueprint\u002F.agents\u002Fskills\u002Fstatus\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":18},"01M75Y31WCYVFM5M4NH7D9MRMW","tests","Set up unit testing with \u002Ftests or \u002Ftests unit, or explicitly set up browser and end-to-end testing with \u002Ftests browser. Reuse or install a suitable test runner, add one example or smoke test, document commands, and verify. Use check for one-time live verification.","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftests\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":18},"01M75Y2ZETQZEHN2V19HWZBB1R","prototype","Plan and create throwaway static HTML and CSS mockups with shared design tokens before feature implementation. Use for \u002Fprototype, screen mockups, layout exploration, themes, or deciding a project's look and feel.","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fprototype\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":18},"01M75Y2WY81F9N40G2RADX2FKH","overview","Validate and normalize project-plan.md and build-plan.md, then generate the durable project-overview.md used by agents. Use for \u002Foverview, plan cleanup, generating the first overview, or refreshing context after either plan changes.","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Foverview\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":18},"01M75Y2SNF3FZJP3NHRT0B5MQE","onboard","Onboard a fresh or early scaffold after Blueprint is overlaid by tuning commands, standards, adapters, visibility, and context loading. Use for \u002Fonboard, fresh installation setup, or what to do after installing Blueprint. Use adopt for an established app.","https:\u002F\u002Fgithub.com\u002Faiblueprinthq\u002Fai-blueprint\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fonboard\u002FSKILL.md",[],[]]