[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7233TC3C68J1JB64GG7J3C9":3,"skill-same:01M7233TC3C68J1JB64GG7J3C9":64},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":33,"ownerHtmlUrl":34,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":35,"install":36,"resources":49},"01M7233TC3C68J1JB64GG7J3C9","askit-build-subagent",null,"Creates and improves Claude subagents (agents\u002F\u003Cname>.md) to the Advanced Skill Library Standard. Use when you need to author a new subagent, scaffold an agents\u002F delegate, declare its tools and chain, or raise an existing subagent's conformance.","product-on-purpose\u002Fagent-skills-toolkit","product-on-purpose","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F253892650?v=4","01M7230Q4442AHBADCVFJCS277","Product on Purpose","01M7230Q4SFDBKY7586D5M7EFX",3,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Faskit-build-subagent\u002FSKILL.md",[],[],"2026-11-10 18:39:20",false,"---\nname: askit-build-subagent\ndescription: Creates and improves Claude subagents (agents\u002F\u003Cname>.md) to the Advanced Skill Library Standard. Use when you need to author a new subagent, scaffold an agents\u002F delegate, declare its tools and chain, or raise an existing subagent's conformance.\nmetadata:\n  version: 0.2.3\n  tier: universal\n  audience: intermediate\n---\n\n# askit-build-subagent\n\n## Purpose\nAuthor conformant subagents. Two modes: `create` scaffolds a new `agents\u002F\u003Cname>.md` (frontmatter + role body); `improve` raises an existing subagent toward the quality bar. Subagents are Claude-only for plugin distribution (Codex v0.135 plugins cannot ship `[agents.*]`; Standard sec 3.3), so this skill emits no Codex artifact. Authoring depth is in [references\u002Fauthoring-subagents.md](references\u002Fauthoring-subagents.md).\n\n## When to use\nWhen the user asks to create, scaffold, write, or improve a subagent (an `agents\u002F\u003Cname>.md` delegate).\n\n## create mode\n1. Brief interview: ask for the subagent name (kebab-case), the bounded job it owns, the narrowest tools it needs, and which components (if any) it may invoke. Skip the interview if these inputs are already in context.\n2. Copy `templates\u002Fagent.md` into `agents\u002F\u003Cname>.md`.\n3. Fill the frontmatter: `name` equal to the file basename; a `description` that states what AND when (Standard sec 8.1); `tools` as the narrowest set (sec 9); optional `model`; `metadata.chain` naming the components it may invoke as a comma-separated string (the recommended shape - a YAML list is silently coerced to a string repr by the reference implementation, see `authoring-chain-contracts.md`; omit `chain` if there is none); `metadata.version`, `metadata.tier`, `metadata.status`, and `metadata.agent-targets: [claude]` (sec 3.3 - subagents are Claude-only for plugin distribution).\n4. Do **not** write `hooks`, `mcpServers`, or `permissionMode`. Claude Code states plainly that \"for security reasons, plugin subagents don't support the `hooks`, `mcpServers`, or `permissionMode` frontmatter fields\" ([subagents documentation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fsub-agents#choose-the-subagent-scope), read 2026-09-28). Writing one does not configure anything; the runtime ignores it, and the author is left believing otherwise. The runtime also ignores `initialPrompt` on a plugin agent ([plugin components](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fplugins\u002Fcomponents#frontmatter-fields-in-plugin-agents)), so leave that out too. The supported set is `name`, `description`, `model`, `effort`, `maxTurns`, `autoCompactWindow`, `tools`, `disallowedTools`, `skills`, `memory`, `background`, `omitClaudeMd`, `isolation` (whose only valid value is `worktree`), `color`, and the `cacheTtl` key of `experimental`. Marketplace scope reports this across every member of a catalogue.\n5. Register the subagent in `library.json` `components.subagents` as `{ name, path, version, tier, status }`.\n6. If the subagent declares a `chain`, add the entry to `agents\u002F_chain-permitted.yaml` (`\u003Cname>: [\u003Cinvoked>, ...]`) so S4 has no orphan.\n7. Assess with `node scripts\u002Fevaluate.mjs . --json` and iterate until 0 errors (S3 components index + S4 chain contract must be clean).\n\n## improve mode\n1. Run `node scripts\u002Fevaluate.mjs . --json` and read the report.\n2. For each finding: an S3 error -> declare\u002Fundeclare the subagent in `components.subagents` to match disk; an S4 orphan -> add the missing `caller -> callee` to `agents\u002F_chain-permitted.yaml`; an S4 phantom (the contract names a component not on disk) -> remove that contract entry, fix a misspelled name, or create the missing component. For any other finding, read its message and apply the fix it states.\n3. Re-run evaluate to confirm.\n\n## Scope\nClaude-only: a subagent is `agents\u002F\u003Cname>.md` (Claude auto-discovers it). Codex does not ingest plugin-shipped subagents (Standard sec 3.3), so there is no Codex render. Authoring is performed inline by the running agent; the toolkit's own `askit-skill-author` subagent is the delegated authoring role for skills.\n","# askit-build-subagent\n\n## Purpose\nAuthor conformant subagents. Two modes: `create` scaffolds a new `agents\u002F\u003Cname>.md` (frontmatter + role body); `improve` raises an existing subagent toward the quality bar. Subagents are Claude-only for plugin distribution (Codex v0.135 plugins cannot ship `[agents.*]`; Standard sec 3.3), so this skill emits no Codex artifact. Authoring depth is in [references\u002Fauthoring-subagents.md](references\u002Fauthoring-subagents.md).\n\n## When to use\nWhen the user asks to create, scaffold, write, or improve a subagent (an `agents\u002F\u003Cname>.md` delegate).\n\n## create mode\n1. Brief interview: ask for the subagent name (kebab-case), the bounded job it owns, the narrowest tools it needs, and which components (if any) it may invoke. Skip the interview if these inputs are already in context.\n2. Copy `templates\u002Fagent.md` into `agents\u002F\u003Cname>.md`.\n3. Fill the frontmatter: `name` equal to the file basename; a `description` that states what AND when (Standard sec 8.1); `tools` as the narrowest set (sec 9); optional `model`; `metadata.chain` naming the components it may invoke as a comma-separated string (the recommended shape - a YAML list is silently coerced to a string repr by the reference implementation, see `authoring-chain-contracts.md`; omit `chain` if there is none); `metadata.version`, `metadata.tier`, `metadata.status`, and `metadata.agent-targets: [claude]` (sec 3.3 - subagents are Claude-only for plugin distribution).\n4. Do **not** write `hooks`, `mcpServers`, or `permissionMode`. Claude Code states plainly that \"for security reasons, plugin subagents don't support the `hooks`, `mcpServers`, or `permissionMode` frontmatter fields\" ([subagents documentation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fsub-agents#choose-the-subagent-scope), read 2026-09-28). Writing one does not configure anything; the runtime ignores it, and the author is left believing otherwise. The runtime also ignores `initialPrompt` on a plugin agent ([plugin components](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fplugins\u002Fcomponents#frontmatter-fields-in-plugin-agents)), so leave that out too. The supported set is `name`, `description`, `model`, `effort`, `maxTurns`, `autoCompactWindow`, `tools`, `disallowedTools`, `skills`, `memory`, `background`, `omitClaudeMd`, `isolation` (whose only valid value is `worktree`), `color`, and the `cacheTtl` key of `experimental`. Marketplace scope reports this across every member of a catalogue.\n5. Register the subagent in `library.json` `components.subagents` as `{ name, path, version, tier, status }`.\n6. If the subagent declares a `chain`, add the entry to `agents\u002F_chain-permitted.yaml` (`\u003Cname>: [\u003Cinvoked>, ...]`) so S4 has no orphan.\n7. Assess with `node scripts\u002Fevaluate.mjs . --json` and iterate until 0 errors (S3 components index + S4 chain contract must be clean).\n\n## improve mode\n1. Run `node scripts\u002Fevaluate.mjs . --json` and read the report.\n2. For each finding: an S3 error -> declare\u002Fundeclare the subagent in `components.subagents` to match disk; an S4 orphan -> add the missing `caller -> callee` to `agents\u002F_chain-permitted.yaml`; an S4 phantom (the contract names a component not on disk) -> remove that contract entry, fix a misspelled name, or create the missing component. For any other finding, read its message and apply the fix it states.\n3. Re-run evaluate to confirm.\n\n## Scope\nClaude-only: a subagent is `agents\u002F\u003Cname>.md` (Claude auto-discovers it). Codex does not ingest plugin-shipped subagents (Standard sec 3.3), so there is no Codex render. Authoring is performed inline by the running agent; the toolkit's own `askit-skill-author` subagent is the delegated authoring role for skills.\n",{"name":5,"metadata":24,"description":7},{"tier":25,"version":26,"audience":27},"universal","0.2.3","intermediate","main","b0a696723b1155d08503a8441c981998f9d13c34","skills\u002Faskit-build-subagent",0,"https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit","https:\u002F\u002Fproduct-on-purpose.github.io\u002Fagent-skills-toolkit","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose","2026-11-14 15:02:57",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add product-on-purpose\u002Fagent-skills-toolkit --skill askit-build-subagent",{"tool":42,"command":43},"bunx","bunx skills add product-on-purpose\u002Fagent-skills-toolkit --skill askit-build-subagent",{"tool":45,"command":46},"pnpm","pnpm dlx skills add product-on-purpose\u002Fagent-skills-toolkit --skill askit-build-subagent","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Faskit-build-subagent\u002FSKILL.md",true,[50,56,60],{"path":51,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":54,"sourceHtmlUrl":55},"examples\u002Fgolden-1-samples-validator-subagent.md","REFERENCE","参考文档",9796,"https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Faskit-build-subagent\u002Fexamples\u002Fgolden-1-samples-validator-subagent.md",{"path":57,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":58,"sourceHtmlUrl":59},"README.md",602,"https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Faskit-build-subagent\u002FREADME.md",{"path":61,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fauthoring-subagents.md",1595,"https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Fskills\u002Faskit-build-subagent\u002Freferences\u002Fauthoring-subagents.md",[65,72,77,82,89,96],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M723BRSVFQ2KRQ8KKY42R6XB","fx-demo","A fixture skill that exists so the re-derived manifest has a component list to compare. Use only from tests\u002Funit\u002Fself-consistency.test.mjs.","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fself-consistency\u002Frepaired\u002Fskills\u002Ffx-demo\u002FSKILL.md",[],[],{"skillNo":73,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M723BPZ1NQDZSEF21D3YP2VF","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fself-consistency\u002Fpre-fix\u002Fskills\u002Ffx-demo\u002FSKILL.md",[],[],{"skillNo":78,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M723BNCZXT3BM577HJYESV08","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fself-consistency\u002Fbody-drift\u002Fskills\u002Ffx-demo\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M723BKWKX9JG10YQ6NK79ZB6","sf-caller","Calls the worker subagent to do a thing. Use when delegating work to sf-worker.","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fgolden\u002Fsubagent-fixture\u002Fskills\u002Fsf-caller\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M723BJNDZHEJTNBJP6ZJA8WM","sf-do-thing","Converts CSV input into a formatted summary. Use when the user asks to summarize or tabulate spreadsheet data.","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fgolden\u002Fsilver-fixture\u002Fskills\u002Fsf-do-thing\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":19},"01M723BH7BYXT0A05PG6QH6B98","root-skill-plugin","Summarizes a changelog section into one release-note sentence. Use when the user asks to condense a list of changelog bullets into a single announcement line.","https:\u002F\u002Fgithub.com\u002Fproduct-on-purpose\u002Fagent-skills-toolkit\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fgolden\u002Froot-skill-plugin\u002FSKILL.md",[],[]]