[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71JKN2RATKA51E0A082MBAF":3,"skill-same:01M71JKN2RATKA51E0A082MBAF":167},{"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":31,"publishedAt":51,"linkOnly":52,"markdownContent":53,"markdownBody":54,"frontMatter":55,"refName":62,"commitSha":63,"skillDirectory":5,"forksCount":64,"repositoryHtmlUrl":65,"repositoryHomepageUrl":66,"ownerHtmlUrl":67,"ownerBlogUrl":68,"ownerDescription":69,"updatedAt":70,"install":71,"resources":84},"01M71JKN2RATKA51E0A082MBAF","nib",null,"Turns an idea or an article into original white-background, hand-drawn editorial illustrations starring a recurring avatar the user owns — one caught scene per idea, in one of several print looks. Use when the user asks to illustrate a post\u002Farticle\u002Fconcept with a consistent character, or invokes \"nib\". Not for generic draw\u002Fmake-an-image requests.","caezium\u002Fskills","caezium","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113233555?v=4","01M71JB77ZPBAWEK009Y2W8BP7","Henry Zhang","01M71JB792NVXCWSYD27HQ0964",3,"MIT","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002FSKILL.md",[18,23,26],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":27,"categoryName":28,"taxonomyType":29,"taxonomyTypeDesc":30},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[5,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50],"when","avatar","post","with","hand","print","one","illustrate","image","URL","explainer","use","for","generic","python3","recurring","concept","editorial","SET","2026-11-10 10:12:30",false,"---\nname: nib\ndescription: >-\n  Turns an idea or an article into original white-background, hand-drawn\n  editorial illustrations starring a recurring avatar the user owns — one\n  caught scene per idea, in one of several print looks. Use when the user asks\n  to illustrate a post\u002Farticle\u002Fconcept with a consistent character, or invokes\n  \"nib\". Not for generic draw\u002Fmake-an-image requests.\nversion: 0.1.1\nargument-hint: \"[idea or article path\u002FURL] | build a character | set avatar \u003Cimage> | style \u003Cname> | explainer\"\nlicense: MIT\nmetadata:\n  category: creative\n  requires:\n    bins: [python3]\n---\n\n# Nib\n\nMake original editorial illustrations for written content. **One image explains\none idea.** A **recurring avatar** — supplied once by the user — is the subject\nof every scene and *performs* the idea (pushes, sorts, steers, builds, holds);\nit is never decoration. The methodology is the constant; the **avatar** and the\n**look** are the parameters.\n\nThis is a configurable house style, not a generic image generator. It is\nintentionally not a photo, not a logo, not a corporate infographic, not a UI\nmockup.\n\n## Setup (once)\n\n1. **A character.** The user's own **avatar image** (a mascot, logo character,\n   anything — save its path); a bundled one via `--avatar-pack \u003Cname>` (see\n   `characters\u002F`); or **build a new one** with the user\n   (`references\u002Fcharacter-builder.md`). The character is the reference on every\n   generation.\n2. **A backend** — one of:\n   - **Codex (free)** — a logged-in Codex CLI (`codex login`). Generates on the\n     user's ChatGPT \u002F Codex subscription, no API key. *Default when no key is set.*\n   - **OpenRouter** — an `OPENROUTER_API_KEY` (`sk-or-…`). Exact 16:9, and lets\n     you pick the model with `--model`.\n\n## Workflow\n\n1. **Read the input.**\n   - A **URL** → fetch the page and extract the readable article text first\n     (use your web-fetch \u002F Defuddle capability), then treat it as an article.\n   - A single **idea** → one illustration.\n   - An **article \u002F post** (or the user's **own draft \u002F notes** they want\n     illustrated) → pick the **load-bearing moments** (a judgment, a flow, a\n     before\u002Fafter, a trap, a loop), 4–8 of them. Not one image per paragraph —\n     the ones that matter. Treat this set as a **starting point the user\n     shapes**: show it first, let them drop moments they don't want, **add\n     their own**, or ask for **more candidates**, then illustrate only the\n     chosen ones. (The desktop app's Article mode exposes the same controls —\n     per-shot select, add-a-shot, \"More ideas\", and a parallel \"Generate\".)\n2. **Pick a look** for the piece — one of **13** (see `references\u002Fstyles.md`):\n   marker, riso, blueprint, woodcut, pixel, clay, gouache, chalk, diorama,\n   enamel, felt, manila, phosphor. One look per piece.\n3. For each idea, **invent a fresh, concrete physical metaphor** in which the\n   avatar performs the idea (see `references\u002Fmethodology.md`). One **caught\n   scene** is the default; switch to the **explainer** register — a hand-drawn\n   sketch-diagram — only when the user asks (\"diagram the flow\") or the idea is\n   itself a traceable structure (`references\u002Fcomposition.md`, `--register explainer`).\n4. **Generate** by calling the engine once per image:\n\n   ```sh\n   python3 scripts\u002Fgenerate.py \\\n     --idea \"trust is built one piece of evidence at a time\" \\\n     --style marker \\\n     --avatar \u002Fpath\u002Fto\u002Favatar.png \\\n     --out .\u002Fout\u002Ftrust.png\n   ```\n\n   It builds the full prompt (methodology + look + idea), sends the character as\n   the reference for consistency, and writes a 16:9 PNG. Useful flags:\n   - `--backend auto|openrouter|codex` — `auto` uses the OpenRouter key if set,\n     else the free Codex lane.\n   - `--model \u003Cid>` — OpenRouter image model (default `google\u002Fgemini-3.1-flash-image-preview`).\n   - `--avatar-pack \u003Cname>` — use a bundled character (see `characters\u002F`) instead\n     of `--avatar`.\n   - `--avatar-spec \"\u003Ctext>\"` — a written description of the character (silhouette,\n     face, the one accent part) that locks its design. Combine with the image —\n     it markedly improves consistency.\n   - `--accent \"\u003Chex|name>\"` — pin the one accent colour (a brand hex \u002F site\n     colour); see `references\u002Fpalettes.md`.\n   - `--register explainer` — draw a hand-built sketch-diagram instead of a single\n     caught scene (`references\u002Fcomposition.md`).\n   - `--doctor` — print backend readiness (which lane will run, and why) and exit.\n5. **Review** against `references\u002Fquality-bar.md` (white background, one idea,\n   avatar performing the action, short labels only, not a slide). Regenerate\n   any that miss.\n6. Report each saved path.\n\n## Example prompts\n\nInvoke Nib from your agent like:\n\n- `Use nib to illustrate: \"one fish, many dishes\"`\n- `Use nib to make illustrations for this article: https:\u002F\u002Fjamesclear.com\u002Ffeedback-loops`\n- `Use nib to build a character with me, then illustrate \"the information well\"`\n- `Use nib to diagram the flow of a RAG pipeline (explainer)`\n- `Use nib to turn this post into a 5-image set in chalk: \u003Cpaste a URL or text>`\n\n## Notes\n\n- Keep text in the image short; the model can misspell long labels.\n- The avatar is the reference on every call — that is what keeps the character\n  consistent across a whole article.\n- For options, generate 2–3 variants of a shot (run the engine a few times with\n  the same idea) and let the user pick.\n- Cost: **free** on the Codex lane (uses the user's ChatGPT \u002F Codex subscription),\n  or a few cents per image on OpenRouter (`google\u002Fgemini-3.1-flash-image-preview`) —\n  a full post is well under a dollar either way. Run `--doctor` to see which lane\n  is ready.\n- A written **`--avatar-spec`** (describe the character in words) markedly improves\n  consistency over the image reference alone — use it whenever you have a description.\n- Inspired by [xiaohei (小黑)](https:\u002F\u002Fgithub.com\u002Fhelloianneo\u002Fian-xiaohei-illustrations)\n  by helloianneo — this methodology is an adaptation of its hand-drawn editorial\n  approach, reworded in Nib's own voice.\n","# Nib\n\nMake original editorial illustrations for written content. **One image explains\none idea.** A **recurring avatar** — supplied once by the user — is the subject\nof every scene and *performs* the idea (pushes, sorts, steers, builds, holds);\nit is never decoration. The methodology is the constant; the **avatar** and the\n**look** are the parameters.\n\nThis is a configurable house style, not a generic image generator. It is\nintentionally not a photo, not a logo, not a corporate infographic, not a UI\nmockup.\n\n## Setup (once)\n\n1. **A character.** The user's own **avatar image** (a mascot, logo character,\n   anything — save its path); a bundled one via `--avatar-pack \u003Cname>` (see\n   `characters\u002F`); or **build a new one** with the user\n   (`references\u002Fcharacter-builder.md`). The character is the reference on every\n   generation.\n2. **A backend** — one of:\n   - **Codex (free)** — a logged-in Codex CLI (`codex login`). Generates on the\n     user's ChatGPT \u002F Codex subscription, no API key. *Default when no key is set.*\n   - **OpenRouter** — an `OPENROUTER_API_KEY` (`sk-or-…`). Exact 16:9, and lets\n     you pick the model with `--model`.\n\n## Workflow\n\n1. **Read the input.**\n   - A **URL** → fetch the page and extract the readable article text first\n     (use your web-fetch \u002F Defuddle capability), then treat it as an article.\n   - A single **idea** → one illustration.\n   - An **article \u002F post** (or the user's **own draft \u002F notes** they want\n     illustrated) → pick the **load-bearing moments** (a judgment, a flow, a\n     before\u002Fafter, a trap, a loop), 4–8 of them. Not one image per paragraph —\n     the ones that matter. Treat this set as a **starting point the user\n     shapes**: show it first, let them drop moments they don't want, **add\n     their own**, or ask for **more candidates**, then illustrate only the\n     chosen ones. (The desktop app's Article mode exposes the same controls —\n     per-shot select, add-a-shot, \"More ideas\", and a parallel \"Generate\".)\n2. **Pick a look** for the piece — one of **13** (see `references\u002Fstyles.md`):\n   marker, riso, blueprint, woodcut, pixel, clay, gouache, chalk, diorama,\n   enamel, felt, manila, phosphor. One look per piece.\n3. For each idea, **invent a fresh, concrete physical metaphor** in which the\n   avatar performs the idea (see `references\u002Fmethodology.md`). One **caught\n   scene** is the default; switch to the **explainer** register — a hand-drawn\n   sketch-diagram — only when the user asks (\"diagram the flow\") or the idea is\n   itself a traceable structure (`references\u002Fcomposition.md`, `--register explainer`).\n4. **Generate** by calling the engine once per image:\n\n   ```sh\n   python3 scripts\u002Fgenerate.py \\\n     --idea \"trust is built one piece of evidence at a time\" \\\n     --style marker \\\n     --avatar \u002Fpath\u002Fto\u002Favatar.png \\\n     --out .\u002Fout\u002Ftrust.png\n   ```\n\n   It builds the full prompt (methodology + look + idea), sends the character as\n   the reference for consistency, and writes a 16:9 PNG. Useful flags:\n   - `--backend auto|openrouter|codex` — `auto` uses the OpenRouter key if set,\n     else the free Codex lane.\n   - `--model \u003Cid>` — OpenRouter image model (default `google\u002Fgemini-3.1-flash-image-preview`).\n   - `--avatar-pack \u003Cname>` — use a bundled character (see `characters\u002F`) instead\n     of `--avatar`.\n   - `--avatar-spec \"\u003Ctext>\"` — a written description of the character (silhouette,\n     face, the one accent part) that locks its design. Combine with the image —\n     it markedly improves consistency.\n   - `--accent \"\u003Chex|name>\"` — pin the one accent colour (a brand hex \u002F site\n     colour); see `references\u002Fpalettes.md`.\n   - `--register explainer` — draw a hand-built sketch-diagram instead of a single\n     caught scene (`references\u002Fcomposition.md`).\n   - `--doctor` — print backend readiness (which lane will run, and why) and exit.\n5. **Review** against `references\u002Fquality-bar.md` (white background, one idea,\n   avatar performing the action, short labels only, not a slide). Regenerate\n   any that miss.\n6. Report each saved path.\n\n## Example prompts\n\nInvoke Nib from your agent like:\n\n- `Use nib to illustrate: \"one fish, many dishes\"`\n- `Use nib to make illustrations for this article: https:\u002F\u002Fjamesclear.com\u002Ffeedback-loops`\n- `Use nib to build a character with me, then illustrate \"the information well\"`\n- `Use nib to diagram the flow of a RAG pipeline (explainer)`\n- `Use nib to turn this post into a 5-image set in chalk: \u003Cpaste a URL or text>`\n\n## Notes\n\n- Keep text in the image short; the model can misspell long labels.\n- The avatar is the reference on every call — that is what keeps the character\n  consistent across a whole article.\n- For options, generate 2–3 variants of a shot (run the engine a few times with\n  the same idea) and let the user pick.\n- Cost: **free** on the Codex lane (uses the user's ChatGPT \u002F Codex subscription),\n  or a few cents per image on OpenRouter (`google\u002Fgemini-3.1-flash-image-preview`) —\n  a full post is well under a dollar either way. Run `--doctor` to see which lane\n  is ready.\n- A written **`--avatar-spec`** (describe the character in words) markedly improves\n  consistency over the image reference alone — use it whenever you have a description.\n- Inspired by [xiaohei (小黑)](https:\u002F\u002Fgithub.com\u002Fhelloianneo\u002Fian-xiaohei-illustrations)\n  by helloianneo — this methodology is an adaptation of its hand-drawn editorial\n  approach, reworded in Nib's own voice.\n",{"name":5,"license":15,"version":56,"metadata":57,"description":7,"argument-hint":61},"0.1.1",{"category":58,"requires":59},"creative",{"bins":60},[46],"[idea or article path\u002FURL] | build a character | set avatar \u003Cimage> | style \u003Cname> | explainer","main","6ef866230c1ba3459fd08fc65141dfd37fffd41a",0,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills","https:\u002F\u002Fskills.henryzh.dev","https:\u002F\u002Fgithub.com\u002Fcaezium","henryzh.dev","quant, data science, full-stack, robotics.\r\nbuilding cool stuff","2026-08-24 19:21:44",{"source":8,"skillName":5,"commands":72,"prompt":82,"zipAvailable":83},[73,76,79],{"tool":74,"command":75},"npx","npx skills add caezium\u002Fskills --skill nib",{"tool":77,"command":78},"bunx","bunx skills add caezium\u002Fskills --skill nib",{"tool":80,"command":81},"pnpm","pnpm dlx skills add caezium\u002Fskills --skill nib","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002FSKILL.md",true,[85,91,97,101,105,109,113,117,121,125,129,133,137,141,145,149,153,157,161],{"path":86,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":89,"sourceHtmlUrl":90},"characters\u002Fblip\u002Favatar.md","REFERENCE","参考文档",398,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fblip\u002Favatar.md",{"path":92,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":95,"sourceHtmlUrl":96},"characters\u002Fblip\u002Freference.png","ASSET","资产文件",185830,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fblip\u002Freference.png",{"path":98,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":99,"sourceHtmlUrl":100},"characters\u002Fhen\u002Favatar.md",264,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fhen\u002Favatar.md",{"path":102,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":103,"sourceHtmlUrl":104},"characters\u002Fhen\u002Freference.png",800679,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fhen\u002Freference.png",{"path":106,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":107,"sourceHtmlUrl":108},"characters\u002Fmo\u002Favatar.md",287,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fmo\u002Favatar.md",{"path":110,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":111,"sourceHtmlUrl":112},"characters\u002Fmo\u002Freference.png",743340,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fmo\u002Freference.png",{"path":114,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":115,"sourceHtmlUrl":116},"characters\u002Fpip\u002Favatar.md",349,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fpip\u002Favatar.md",{"path":118,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":119,"sourceHtmlUrl":120},"characters\u002Fpip\u002Freference.png",733382,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fpip\u002Freference.png",{"path":122,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":123,"sourceHtmlUrl":124},"characters\u002FREADME.md",1551,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002FREADME.md",{"path":126,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":127,"sourceHtmlUrl":128},"characters\u002Fscout\u002Favatar.md",278,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fscout\u002Favatar.md",{"path":130,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":131,"sourceHtmlUrl":132},"characters\u002Fscout\u002Freference.png",334164,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fcharacters\u002Fscout\u002Freference.png",{"path":134,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":135,"sourceHtmlUrl":136},"references\u002Fcharacter-builder.md",3725,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fcharacter-builder.md",{"path":138,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":139,"sourceHtmlUrl":140},"references\u002Fcomposition.md",4176,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fcomposition.md",{"path":142,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":143,"sourceHtmlUrl":144},"references\u002Fmethodology.md",1641,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fmethodology.md",{"path":146,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":147,"sourceHtmlUrl":148},"references\u002Fpalettes.md",2213,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fpalettes.md",{"path":150,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":151,"sourceHtmlUrl":152},"references\u002Fquality-bar.md",2258,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fquality-bar.md",{"path":154,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":155,"sourceHtmlUrl":156},"references\u002Fstyle-data.json",9923,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fstyle-data.json",{"path":158,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":159,"sourceHtmlUrl":160},"references\u002Fstyles.md",1431,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Freferences\u002Fstyles.md",{"path":162,"resourceType":163,"resourceTypeDesc":164,"sizeBytes":165,"sourceHtmlUrl":166},"scripts\u002Fgenerate.py","SCRIPT","脚本",16026,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fnib\u002Fscripts\u002Fgenerate.py",[168,181,209,227,247,269],{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":176,"publishedAt":51},"01M71JT4RYCAS18RF6003YYR1D","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwriting-for-agents\u002FSKILL.md",[174,175],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":27,"categoryName":28,"taxonomyType":29,"taxonomyTypeDesc":30},[177,44,178,179,43,180,32],"writing for agents","writing","editing","documents",{"skillNo":182,"name":183,"nameZh":6,"description":184,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":185,"categories":186,"keywords":190,"publishedAt":51},"01M71JT0GES4AZ5ADH8EADNN1Z","wrangler","Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store. Load before running wrangler commands to ensure correct syntax and best practices. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwrangler\u002FSKILL.md",[187],{"categoryNo":188,"categoryName":189,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[183,191,192,193,194,195,196,197,198,199,200,201,202,44,203,204,205,206,207,208],"cloudflare","d1","knowledge","workflows","cli","secrets","commands","syntax","cloudflare workers","r2","queues","running","containers","correct","load","workers","best practices","store",{"skillNo":210,"name":211,"nameZh":6,"description":212,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":213,"categories":214,"keywords":215,"publishedAt":51},"01M71JSWX4R203M32K9A18AAW8","workers-best-practices","Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for common Workers anti-patterns (streaming, floating promises, global state, secrets, bindings, observability). Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fworkers-best-practices\u002FSKILL.md",[],[216,206,207,217,218,219,44,191,220,193,221,183,178,205,32,222,223,224,225,226],"workers best practices","observability","reviews","promises","worker","new","code","common","streaming","state","anti-patterns",{"skillNo":228,"name":229,"nameZh":6,"description":230,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":231,"categories":232,"keywords":233,"publishedAt":51},"01M71JST6P88MXGT2MWDTPJGEG","wizard","Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwizard\u002FSKILL.md",[],[229,202,234,44,235,236,237,238,43,239,32,240,38,241,242,243,244,245,246,196],"setting","up","ci","provisioning infrastructure","only","invoke","Bash","interactive","can","credentials","dashboard","generate","human",{"skillNo":248,"name":249,"nameZh":6,"description":250,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":251,"sourceHtmlUrl":252,"categories":253,"keywords":254,"publishedAt":51},"01M71JSQ95S7BQ61DAPN95X2W4","why","Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available evidence sources, searches each relevant category, and returns a cited read on decisions and tradeoffs. Use how for runtime behavior.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FSKILL.md",[],[249,255,256,257,258,44,259,260,43,261,262,263,264,265,266,267,268],"sources","thresholds","evidence","relevant","runtime","design","decisions","work","how","behavior","read","data","rationale","returns",{"skillNo":270,"name":271,"nameZh":6,"description":272,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":273,"categories":274,"keywords":278,"publishedAt":51},"01M71JSMHFVSH8HK8GXN6YAX6J","week-prep","Generate a briefing with calendar, tasks, and attention flags. Use for \"week prep\", \"briefing\", or \"what's on this week\".","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fweek-prep\u002FSKILL.md",[275],{"categoryNo":276,"categoryName":277,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[279,280,281,282,283,44,284,285,286,287,43,240,288,35,289,245,290,265,291,292],"week prep","prep","deep","attention","get","search","messages","briefing","MCP","tasks","slack","calendar","flags","grep"]