[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CNBWKJ6RSY41DGWS38YY0Y":3,"skill-same:01M6CNBWKJ6RSY41DGWS38YY0Y":88},{"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":23,"publishedAt":35,"linkOnly":36,"markdownContent":37,"markdownBody":38,"frontMatter":39,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":53,"repositoryHtmlUrl":54,"repositoryHomepageUrl":55,"ownerHtmlUrl":56,"ownerBlogUrl":57,"ownerDescription":58,"updatedAt":59,"install":60,"resources":73},"01M6CNBWKJ6RSY41DGWS38YY0Y","skill-builder",null,"Meta-skill for creating new skills from natural language. Trigger phrases: \"new skill\", \"create a skill\", \"build a skill\", \"make a skill\", \"generate a skill\", \"author a skill\", \"skill builder\".","anton-abyzov\u002Fvskill","anton-abyzov","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F979819?v=4","01M6CN5XDNZDM4A59WYJ1MF0DF","Anton Abyzov","01M6CN5XHVS7FXP067EJQVVZHK",47,"MIT","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fplugins\u002Fskills\u002Fskills\u002Fskill-builder\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34],"Skill Builder","builder","plugin","create","generate","https","Meta","build","language","new skill","natural language","2026-11-10 10:12:30",false,"---\nname: skill-builder\ndescription: >\n  Meta-skill for creating new skills from natural language. Trigger phrases:\n  \"new skill\", \"create a skill\", \"build a skill\", \"make a skill\",\n  \"generate a skill\", \"author a skill\", \"skill builder\".\nversion: 1.0.5\nmetadata:\n  version: 1.0.4\n  homepage: https:\u002F\u002Fverified-skill.com\u002Fskills\u002Fskill-builder\n  tags:\n    - skill-authoring\n    - plugin-authoring\n    - meta\n    - universal\n---\n\n# skill-builder\n\nAuthor new universal skills from natural-language prompts across any agent\nhost (Claude Code, Cursor, Codex, Gemini CLI, Windsurf, OpenCode, and the\nrest of the 50+ agents in the vskill registry).\n\n## What this skill does\n\n`skill-builder` is a cross-tool authoring meta-skill. When the host agent\nsees a trigger phrase (see frontmatter `description`), this skill takes\nover and produces a new SKILL.md — emitted for every target agent, not\njust the one currently invoking it. The emitted skills carry an\n`x-sw-schema-version: 1` marker so future tooling can evolve the format\nwithout breaking older installs.\n\nIt does this by delegating to the real generator via a three-path\nfallback chain:\n\n- **Path A — `vskill skill` CLI** (preferred). The CLI owns the full\n  authoring pipeline: prompt parsing, target-agent lowering, divergence\n  report, security scan, and disk write.\n- **Path B — vskill Studio (browser)**. When the CLI is not on PATH but\n  the npm package is present (e.g. a project-local install), start the\n  eval server and walk the user through the Studio UI.\n- **Path C — Anthropic `skill-creator`** (Claude Code only fallback).\n  Delegate to the built-in Anthropic authoring skill. This produces a\n  single Claude-flavored SKILL.md with no universal targets, so we log\n  the degradation so users know what they're getting.\n\nEach path is tried in order; the first one whose detection check\nsucceeds wins. If none match, we emit a remediation message and stop.\n\n## Detection script\n\nThe host agent runs this snippet (or the equivalent) before authoring\nanything. Return on the first path whose probe exits 0.\n\n```bash\n# Path A: vskill CLI on PATH\nif which vskill >\u002Fdev\u002Fnull 2>&1; then\n  # Invoke the CLI. --targets is optional; default is the full\n  # universal set. --prompt is the user's natural-language request.\n  vskill skill new --prompt \"\u003CUSER_PROMPT>\" [--targets=\u003Ccsv>]\n  exit $?\nfi\n\n# Path B: vskill package resolvable but no CLI on PATH\nif node -e \"require.resolve('vskill')\" >\u002Fdev\u002Fnull 2>&1; then\n  vskill eval serve\n  echo \"[skill-builder] open the URL printed above and use the Studio 'New skill' flow\"\n  exit 0\nfi\n\n# Path C: Claude Code host with Anthropic skill-creator installed\nif [ \"${SW_HOST_AGENT:-claude-code}\" = \"claude-code\" ] \\\n   && test -d \"$HOME\u002F.claude\u002Fskills\u002Fskill-creator\"; then\n  echo \"[skill-builder] fallback mode — universal targets not emitted; install vskill for universal support\"\n  # Delegate to the Anthropic built-in. The host agent will route\n  # the original user prompt into skill-creator.\n  exit 0\nfi\n\n# Nothing matched — print remediation and fail loudly.\ncat \u003C\u003C'EOF'\n[skill-builder] no authoring backend available. Install one of:\n\n  npm install -g vskill          # preferred — universal targets\n  claude plugin install skill-creator   # Claude Code only, no universal emit\n\nEOF\nexit 1\n```\n\nNotes for the host agent:\n\n- `\u003CUSER_PROMPT>` is the user's raw natural-language request.\n- `\u003Ccsv>` for `--targets` is a comma-separated list of agent IDs\n  (e.g. `claude-code,codex,cursor`). Omit to default to the full\n  universal set.\n- `SW_HOST_AGENT` is the environment hint set by the invoking tool.\n  When unset, assume `claude-code` — that is the only host that ships\n  `skill-creator` as a built-in, so the default is safe.\n- The log line in Path C is intentionally exact; test gates grep for\n  it to confirm fallback activation.\n\n## Sentinel file contract\n\nPath A MUST write a sentinel file to the current working directory so\ntest gates and manual verification can confirm activation without\nparsing CLI stdout. The file is:\n\n```\n.skill-builder-invoked.json\n```\n\nWith shape:\n\n```json\n{\n  \"trigger\": \"new skill\",\n  \"agent\": \"claude-code\",\n  \"timestamp\": \"2026-04-24T14:35:00.000Z\",\n  \"targets\": [\"claude-code\", \"codex\", \"cursor\"],\n  \"prompt\": \"lint markdown files for broken links\"\n}\n```\n\nField definitions:\n\n- `trigger` — the exact phrase that activated `skill-builder`\n  (e.g. `\"new skill\"`, `\"create a skill\"`).\n- `agent` — the host agent id that invoked us\n  (`claude-code`, `cursor`, `codex`, etc.).\n- `timestamp` — ISO-8601 UTC timestamp at invocation.\n- `targets` — the resolved target agent IDs the CLI will emit for.\n  Defaults to the full universal set when the user did not pass\n  `--targets`.\n- `prompt` — the raw user prompt passed to the CLI.\n\nPaths B and C do NOT write the sentinel — they have their own\nverification surfaces (Studio UI state for B, Claude Code's native\nskill-creator log for C). Consumers that need \"did skill-builder\nactivate?\" gating must check the sentinel AND tolerate its absence\nin fallback modes.\n\n## Authoring a plugin (multi-skill bundle)\n\nClaude Code's plugin system lets you ship 2+ related skills under a single\nshared identity (a folder with `.claude-plugin\u002Fplugin.json` and one or more\n`skills\u002F\u003Cslug>\u002FSKILL.md` underneath). Prefer plugin mode over standalone\nwhen **the user describes 2+ related skills under a shared identity** —\ne.g. \"a plugin called `hi-anton` with `greet` and `farewell` skills\", or\n\"bundle these under one namespace for distribution\".\n\n`claude plugin` CLI has no `new`\u002F`create` subcommand — plugin authoring is\notherwise unowned. There are three concrete paths to land a valid plugin\nmanifest, all of which delegate schema validation to\n`claude plugin validate \u003Cpath>` so vskill never duplicates Claude Code's\nplugin schema:\n\n1. **`vskill plugin new \u003Cname> [--with-skill \u003Cslug>] [--description \"...\"]`**\n   *(CLI; preferred when working from the terminal)*. Scaffolds\n   `\u003Cname>\u002F.claude-plugin\u002Fplugin.json` and (optionally) a first skill.\n   Runs `claude plugin validate \u003Cname>` after writing; on failure the\n   manifest is unlinked so a broken plugin never lands.\n\n2. **Studio Create-Skill modal → \"Plugin (multi-skill)\" mode** (writes both\n   `\u003Cplugin>\u002F.claude-plugin\u002Fplugin.json` and a first\n   `\u003Cplugin>\u002Fskills\u002F\u003Cskill>\u002FSKILL.md` in a single click).\n\n3. **Studio sidebar \"Convert →\" CTA** *(when standalone skills already\n   exist)*. The AUTHORING > Skills section detects folders with 2+\n   standalone skills and offers a one-click promotion to a real plugin.\n   Behind the scenes this calls `POST \u002Fapi\u002Fauthoring\u002Fconvert-to-plugin`.\n\nFolder shape after authoring:\n\n```\nmy-plugin\u002F\n├── .claude-plugin\u002F\n│   └── plugin.json          # {\"name\":\"my-plugin\",\"description\":\"…\"} — author\u002Fversion derived at install time\n└── skills\u002F\n    ├── greet\u002FSKILL.md\n    └── farewell\u002FSKILL.md\n```\n\nOnce any of the three paths writes a valid manifest, the next refresh in\nSkill Studio promotes the affected skills' `scopeV2` from\n`authoring-project` to `authoring-plugin` automatically — no further user\naction needed.\n\n## What this is NOT\n\n- **Not `scout`** — `scout` is for DISCOVERING skills that already\n  exist on the verified-skill.com registry (\"find me a kubernetes\n  skill\"). `skill-builder` is for CREATING new skills from a prompt.\n  If the user's intent is \"find\" or \"install\" rather than \"make\",\n  route to `scout` instead.\n- **Not `sw:skill-gen`** — `sw:skill-gen` is the SpecWeave command\n  for signal-based skill GENERATION: it reads patterns from living\n  docs and emits project-specific skills. That flow is automatic\n  and pattern-driven, whereas `skill-builder` is interactive and\n  prompt-driven. Use `sw:skill-gen` inside a SpecWeave project when\n  you want skills derived from detected code patterns.\n\nIn short:\n\n| Intent                               | Skill              |\n|--------------------------------------|--------------------|\n| \"find me a skill for X\"              | `scout`            |\n| \"generate skills from my codebase\"   | `sw:skill-gen`     |\n| \"make a new skill from this prompt\"  | `skill-builder`    |\n\n## References\n\n- [`references\u002Ftarget-agents.md`](references\u002Ftarget-agents.md) —\n  the full target agent table with universality flags and local\n  skills dirs, sourced from `src\u002Fagents\u002Fagents-registry.ts`.\n- [`references\u002Fdivergence-report-schema.md`](references\u002Fdivergence-report-schema.md) —\n  the format spec for per-skill `\u003Cname>-divergence.md` files that\n  document frontmatter fields dropped or translated per target.\n- [`references\u002Ffallback-modes.md`](references\u002Ffallback-modes.md) —\n  the exact detection commands and exact warning messages for the\n  A → B → C transitions, mirrored from this file so the two stay\n  in sync.\n","# skill-builder\n\nAuthor new universal skills from natural-language prompts across any agent\nhost (Claude Code, Cursor, Codex, Gemini CLI, Windsurf, OpenCode, and the\nrest of the 50+ agents in the vskill registry).\n\n## What this skill does\n\n`skill-builder` is a cross-tool authoring meta-skill. When the host agent\nsees a trigger phrase (see frontmatter `description`), this skill takes\nover and produces a new SKILL.md — emitted for every target agent, not\njust the one currently invoking it. The emitted skills carry an\n`x-sw-schema-version: 1` marker so future tooling can evolve the format\nwithout breaking older installs.\n\nIt does this by delegating to the real generator via a three-path\nfallback chain:\n\n- **Path A — `vskill skill` CLI** (preferred). The CLI owns the full\n  authoring pipeline: prompt parsing, target-agent lowering, divergence\n  report, security scan, and disk write.\n- **Path B — vskill Studio (browser)**. When the CLI is not on PATH but\n  the npm package is present (e.g. a project-local install), start the\n  eval server and walk the user through the Studio UI.\n- **Path C — Anthropic `skill-creator`** (Claude Code only fallback).\n  Delegate to the built-in Anthropic authoring skill. This produces a\n  single Claude-flavored SKILL.md with no universal targets, so we log\n  the degradation so users know what they're getting.\n\nEach path is tried in order; the first one whose detection check\nsucceeds wins. If none match, we emit a remediation message and stop.\n\n## Detection script\n\nThe host agent runs this snippet (or the equivalent) before authoring\nanything. Return on the first path whose probe exits 0.\n\n```bash\n# Path A: vskill CLI on PATH\nif which vskill >\u002Fdev\u002Fnull 2>&1; then\n  # Invoke the CLI. --targets is optional; default is the full\n  # universal set. --prompt is the user's natural-language request.\n  vskill skill new --prompt \"\u003CUSER_PROMPT>\" [--targets=\u003Ccsv>]\n  exit $?\nfi\n\n# Path B: vskill package resolvable but no CLI on PATH\nif node -e \"require.resolve('vskill')\" >\u002Fdev\u002Fnull 2>&1; then\n  vskill eval serve\n  echo \"[skill-builder] open the URL printed above and use the Studio 'New skill' flow\"\n  exit 0\nfi\n\n# Path C: Claude Code host with Anthropic skill-creator installed\nif [ \"${SW_HOST_AGENT:-claude-code}\" = \"claude-code\" ] \\\n   && test -d \"$HOME\u002F.claude\u002Fskills\u002Fskill-creator\"; then\n  echo \"[skill-builder] fallback mode — universal targets not emitted; install vskill for universal support\"\n  # Delegate to the Anthropic built-in. The host agent will route\n  # the original user prompt into skill-creator.\n  exit 0\nfi\n\n# Nothing matched — print remediation and fail loudly.\ncat \u003C\u003C'EOF'\n[skill-builder] no authoring backend available. Install one of:\n\n  npm install -g vskill          # preferred — universal targets\n  claude plugin install skill-creator   # Claude Code only, no universal emit\n\nEOF\nexit 1\n```\n\nNotes for the host agent:\n\n- `\u003CUSER_PROMPT>` is the user's raw natural-language request.\n- `\u003Ccsv>` for `--targets` is a comma-separated list of agent IDs\n  (e.g. `claude-code,codex,cursor`). Omit to default to the full\n  universal set.\n- `SW_HOST_AGENT` is the environment hint set by the invoking tool.\n  When unset, assume `claude-code` — that is the only host that ships\n  `skill-creator` as a built-in, so the default is safe.\n- The log line in Path C is intentionally exact; test gates grep for\n  it to confirm fallback activation.\n\n## Sentinel file contract\n\nPath A MUST write a sentinel file to the current working directory so\ntest gates and manual verification can confirm activation without\nparsing CLI stdout. The file is:\n\n```\n.skill-builder-invoked.json\n```\n\nWith shape:\n\n```json\n{\n  \"trigger\": \"new skill\",\n  \"agent\": \"claude-code\",\n  \"timestamp\": \"2026-04-24T14:35:00.000Z\",\n  \"targets\": [\"claude-code\", \"codex\", \"cursor\"],\n  \"prompt\": \"lint markdown files for broken links\"\n}\n```\n\nField definitions:\n\n- `trigger` — the exact phrase that activated `skill-builder`\n  (e.g. `\"new skill\"`, `\"create a skill\"`).\n- `agent` — the host agent id that invoked us\n  (`claude-code`, `cursor`, `codex`, etc.).\n- `timestamp` — ISO-8601 UTC timestamp at invocation.\n- `targets` — the resolved target agent IDs the CLI will emit for.\n  Defaults to the full universal set when the user did not pass\n  `--targets`.\n- `prompt` — the raw user prompt passed to the CLI.\n\nPaths B and C do NOT write the sentinel — they have their own\nverification surfaces (Studio UI state for B, Claude Code's native\nskill-creator log for C). Consumers that need \"did skill-builder\nactivate?\" gating must check the sentinel AND tolerate its absence\nin fallback modes.\n\n## Authoring a plugin (multi-skill bundle)\n\nClaude Code's plugin system lets you ship 2+ related skills under a single\nshared identity (a folder with `.claude-plugin\u002Fplugin.json` and one or more\n`skills\u002F\u003Cslug>\u002FSKILL.md` underneath). Prefer plugin mode over standalone\nwhen **the user describes 2+ related skills under a shared identity** —\ne.g. \"a plugin called `hi-anton` with `greet` and `farewell` skills\", or\n\"bundle these under one namespace for distribution\".\n\n`claude plugin` CLI has no `new`\u002F`create` subcommand — plugin authoring is\notherwise unowned. There are three concrete paths to land a valid plugin\nmanifest, all of which delegate schema validation to\n`claude plugin validate \u003Cpath>` so vskill never duplicates Claude Code's\nplugin schema:\n\n1. **`vskill plugin new \u003Cname> [--with-skill \u003Cslug>] [--description \"...\"]`**\n   *(CLI; preferred when working from the terminal)*. Scaffolds\n   `\u003Cname>\u002F.claude-plugin\u002Fplugin.json` and (optionally) a first skill.\n   Runs `claude plugin validate \u003Cname>` after writing; on failure the\n   manifest is unlinked so a broken plugin never lands.\n\n2. **Studio Create-Skill modal → \"Plugin (multi-skill)\" mode** (writes both\n   `\u003Cplugin>\u002F.claude-plugin\u002Fplugin.json` and a first\n   `\u003Cplugin>\u002Fskills\u002F\u003Cskill>\u002FSKILL.md` in a single click).\n\n3. **Studio sidebar \"Convert →\" CTA** *(when standalone skills already\n   exist)*. The AUTHORING > Skills section detects folders with 2+\n   standalone skills and offers a one-click promotion to a real plugin.\n   Behind the scenes this calls `POST \u002Fapi\u002Fauthoring\u002Fconvert-to-plugin`.\n\nFolder shape after authoring:\n\n```\nmy-plugin\u002F\n├── .claude-plugin\u002F\n│   └── plugin.json          # {\"name\":\"my-plugin\",\"description\":\"…\"} — author\u002Fversion derived at install time\n└── skills\u002F\n    ├── greet\u002FSKILL.md\n    └── farewell\u002FSKILL.md\n```\n\nOnce any of the three paths writes a valid manifest, the next refresh in\nSkill Studio promotes the affected skills' `scopeV2` from\n`authoring-project` to `authoring-plugin` automatically — no further user\naction needed.\n\n## What this is NOT\n\n- **Not `scout`** — `scout` is for DISCOVERING skills that already\n  exist on the verified-skill.com registry (\"find me a kubernetes\n  skill\"). `skill-builder` is for CREATING new skills from a prompt.\n  If the user's intent is \"find\" or \"install\" rather than \"make\",\n  route to `scout` instead.\n- **Not `sw:skill-gen`** — `sw:skill-gen` is the SpecWeave command\n  for signal-based skill GENERATION: it reads patterns from living\n  docs and emits project-specific skills. That flow is automatic\n  and pattern-driven, whereas `skill-builder` is interactive and\n  prompt-driven. Use `sw:skill-gen` inside a SpecWeave project when\n  you want skills derived from detected code patterns.\n\nIn short:\n\n| Intent                               | Skill              |\n|--------------------------------------|--------------------|\n| \"find me a skill for X\"              | `scout`            |\n| \"generate skills from my codebase\"   | `sw:skill-gen`     |\n| \"make a new skill from this prompt\"  | `skill-builder`    |\n\n## References\n\n- [`references\u002Ftarget-agents.md`](references\u002Ftarget-agents.md) —\n  the full target agent table with universality flags and local\n  skills dirs, sourced from `src\u002Fagents\u002Fagents-registry.ts`.\n- [`references\u002Fdivergence-report-schema.md`](references\u002Fdivergence-report-schema.md) —\n  the format spec for per-skill `\u003Cname>-divergence.md` files that\n  document frontmatter fields dropped or translated per target.\n- [`references\u002Ffallback-modes.md`](references\u002Ffallback-modes.md) —\n  the exact detection commands and exact warning messages for the\n  A → B → C transitions, mirrored from this file so the two stay\n  in sync.\n",{"name":5,"version":40,"metadata":41,"description":49},"1.0.5",{"tags":42,"version":47,"homepage":48},[43,44,45,46],"skill-authoring","plugin-authoring","meta","universal","1.0.4","https:\u002F\u002Fverified-skill.com\u002Fskills\u002Fskill-builder","Meta-skill for creating new skills from natural language. Trigger phrases: \"new skill\", \"create a skill\", \"build a skill\", \"make a skill\", \"generate a skill\", \"author a skill\", \"skill builder\".\n","main","b198d63224f1ecec8bca1300fb21dfa996c853db","plugins\u002Fskills\u002Fskills\u002Fskill-builder",5,"https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill","https:\u002F\u002Fverified-skill.com","https:\u002F\u002Fgithub.com\u002Fanton-abyzov","https:\u002F\u002Fspecwithcode.com","Founder of SpecWithCode · Creator of SpecWeave · AI coding agents, Cloud, DevOps · 20 yrs SWE · YouTube: @AntonAbyzov","2026-11-13 00:22:51",{"source":8,"skillName":5,"commands":61,"prompt":71,"zipAvailable":72},[62,65,68],{"tool":63,"command":64},"npx","npx skills add anton-abyzov\u002Fvskill --skill skill-builder",{"tool":66,"command":67},"bunx","bunx skills add anton-abyzov\u002Fvskill --skill skill-builder",{"tool":69,"command":70},"pnpm","pnpm dlx skills add anton-abyzov\u002Fvskill --skill skill-builder","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fplugins\u002Fskills\u002Fskills\u002Fskill-builder\u002FSKILL.md",true,[74,80,84],{"path":75,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fdivergence-report-schema.md","REFERENCE","参考文档",2382,"https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fplugins\u002Fskills\u002Fskills\u002Fskill-builder\u002Freferences\u002Fdivergence-report-schema.md",{"path":81,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Ffallback-modes.md",1883,"https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fplugins\u002Fskills\u002Fskills\u002Fskill-builder\u002Freferences\u002Ffallback-modes.md",{"path":85,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Ftarget-agents.md",3391,"https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fplugins\u002Fskills\u002Fskills\u002Fskill-builder\u002Freferences\u002Ftarget-agents.md",[89,116,133,161,197,223],{"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":35},"01M6HKP9JT8CH33W7DJB544DCS","x-reply-drafts","Find live X threads where a product genuinely fits, and draft truthful, humanized replies for the owner to review and post by hand. Human pace only: max 5 replies a day, never two within 20 minutes, no automated posting. Use when the user asks to 'find threads to reply to', 'draft X replies', 'reply-guy session', 'join trending conversations', 'engage on Twitter about \u003Cproduct>', or wants a daily batch of reply drafts. Not for scheduling original posts (social-media-posting) or bulk publishing (blotato).","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fplugins\u002Fmarketing\u002Fskills\u002Fx-reply-drafts\u002FSKILL.md",[],[96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115],"reply","drafts","x reply drafts","review","product","batch","Media","unslop","never","only","humanize","social","human","join","trending","bulk","posts","threads","conversations","hand",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":122,"publishedAt":35},"01M6CND903Q6E6R2XDWMFH8MAY","hello-world","A tiny test skill used by the 0853 verify-runtime harness. Says hello. No external calls. Use when the user asks for a sanity-check skill in a verify-harness fixture.","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Ftest\u002Fverify\u002Ffixtures\u002Ftiny-skill-source\u002Fplugins\u002Ftiny-test-plugin\u002Fskills\u002Fhello-world\u002FSKILL.md",[],[123,124,125,126,127,128,129,130,131,132],"hello world","hello","world","test","Test Skill","harness","verify","check","user","use",{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":147,"publishedAt":35},"01M6CND60H0N1CYEB6ABEW18NS","existing-skill","Lint markdown files for style and grammar issues.","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fsrc\u002Fcommands\u002F__tests__\u002Ffixtures\u002Fexisting-skill\u002FSKILL.md",[139,142],{"categoryNo":140,"categoryName":141,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000002","文档处理",{"categoryNo":143,"categoryName":144,"taxonomyType":145,"taxonomyTypeDesc":146},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[148,149,150,151,152,153,154,155,156,157,158,159,160],"existing skill","docs","lint","code","issues","markdown","grammar","read","write","edit","cursor","files","lint markdown files",{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":177,"publishedAt":35},"01M6CND1JAW97FY3W50MAPNKX7","remotion-best-practices","Best practices for Remotion video creation in React — scaffolding, compositions, animations, transitions — PLUS hard-won rules for high-fidelity product-demo videos: real logos\u002Ficons, precise YouTube-reference matching, full-screen camera zooms, click-targets that actually hit, content-tight rectangle highlights, voiceover regeneration discipline, audio-level normalization (no double-boost), per-scene voice-sync math, Studio detail-page mirror patterns, bold-not-subtle connection lines, and the YouTube publication kit (chapters, description, thumbnail, hashtags) that turns a render into a shipped video.","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fskills\u002Fremotion-best-practices\u002FSKILL.md",[167,170,173,174],{"categoryNo":168,"categoryName":169,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":171,"categoryName":172,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":175,"categoryName":176,"taxonomyType":145,"taxonomyTypeDesc":146},"01JBSEEDCAT000000000000031","产品经理",[178,179,180,181,182,100,183,184,185,186,187,188,189,190,191,192,193,194,195,196],"best practices","remotion best practices","Remotion","react","sync","normalization","Video","Animation","Composition","Audio","page","math","discipline","publishing","camera","youtube","voiceover","patterns","remotion, video, react, animation, composition",{"skillNo":198,"name":199,"nameZh":6,"description":200,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":201,"categories":202,"keywords":204,"publishedAt":35},"01M6CNCXMSB1MKQ7646AR12VXY","rea","Find out exactly how a feature is implemented in a shipped tool or app you don't have the source for (desktop apps, Electron apps like Claude desktop, native CLIs like Codex, websites, APKs, .NET assemblies) using REA - Reverse Engineer Anything (github.com\u002Fmorluto\u002Frea). Sets REA up when it is missing (CLI, MCP registration, Hopper or Ghidra), routes the target to the right REA tool, keeps evidence out of the context window, and turns findings into a clean-room spec before any rebuild. Use when the user says 'use REA', 'reverse engineer', 'how does \u003Capp> do X', 'how is this feature implemented in \u003Ctool>', 'clone this feature', 'decompile', 'look inside the app', or asks how a closed-source competitor works. Skip it when the source is available - read the repo instead.","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fskills\u002Frea\u002FSKILL.md",[203],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[199,205,206,207,208,209,210,211,212,213,155,214,215,216,217,218,219,220,221,222],"MCP","cli","reverse-engineering","ghidra","binary-analysis","analysis","how","reverse engineer anything","repo","context","evidence","up","engineer","rebuild","clean","registration","do","reverse engineer",{"skillNo":224,"name":225,"nameZh":6,"description":226,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":227,"categories":228,"keywords":237,"publishedAt":35},"01M6CNCSE0847Q1FP1GRQ75SR9","motion-graphics-promo","After Effects-style motion graphics for short product promos (15-45 s, vertical or landscape), built as code with HyperFrames + GSAP. Covers a story-first brief, motion grammar (energetic entry, one transition per task, spring settle, 2-3 s of stable proof), a 32-second promo timeline, device-frame and kinetic-type recipes, vertical-first composition, evidence rules (real UI only, no fake clicks), music licensing and encoded-frame verification. Use when the user asks for an After Effects-like animation, motion graphics, a kinetic-type promo, an app promo video, a product teaser, a Reels\u002FTikTok\u002FShorts ad, or 'animations like After Effects'. Also covers short explainers, tutorials and 15-second vertical ads: a design kit (footage beds, condensed caps in a marker box, white device frames, falling stickers, voice-over with music ducking) and HyperFrames render traps that silently break output. Pairs with hyperframes-best-practices (screencast pipeline) and remotion-best-practices.","https:\u002F\u002Fgithub.com\u002Fanton-abyzov\u002Fvskill\u002Fblob\u002Fmain\u002Fskills\u002Fmotion-graphics-promo\u002FSKILL.md",[229,232,233,234],{"categoryNo":230,"categoryName":231,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":171,"categoryName":172,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":143,"categoryName":144,"taxonomyType":145,"taxonomyTypeDesc":146},{"categoryNo":235,"categoryName":236,"taxonomyType":145,"taxonomyTypeDesc":146},"01JBSEEDCAT000000000000032","设计师",[238,239,240,241,242,151,243,100,244,245,246,184,185,180,186,247,154,248,105,249],"graphics","motion graphics","motion","motion graphics promo","design","verification","ui","brief","pipeline","ads","short","tiktok"]