[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C1JS246128J3YFC8AV39GY":3,"skill-same:01M6C1JS246128J3YFC8AV39GY":92},{"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":55,"linkOnly":56,"markdownContent":57,"markdownBody":58,"frontMatter":59,"refName":69,"commitSha":70,"skillDirectory":71,"forksCount":72,"repositoryHtmlUrl":73,"repositoryHomepageUrl":74,"ownerHtmlUrl":75,"ownerBlogUrl":76,"ownerDescription":77,"updatedAt":78,"install":79,"resources":91},"01M6C1JS246128J3YFC8AV39GY","pipeline-design",null,"Interactive LLM inference pipeline design — elicits requirements, recommends pattern, scaffolds production-ready artifacts","jmagly\u002Faiwg","jmagly","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1159087?v=4","01M6C10AK8N2WCSQN11FYGTJWN","Joseph Magly","01M6C10AQ8A2RMED1Y1BSVY7NF",229,"MIT","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Faddons\u002Fnlp-prod\u002Fskills\u002Fpipeline-design\u002FSKILL.md",[18,23,26],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":27,"categoryName":28,"taxonomyType":29,"taxonomyTypeDesc":30},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54],"pipeline","pipeline design","design","write","production-ready","true","requirements","efficiency","volume","pattern","ready","use","webfetch","interactive","read","llm inference","haiku","Production","artifacts","inference","python","nlp","typescript","2026-11-10 10:12:30",false,"---\nnamespace: aiwg\nname: pipeline-design\nplatforms: [all]\ndescription: Interactive LLM inference pipeline design — elicits requirements, recommends pattern, scaffolds production-ready artifacts\ncommandHint:\n  argumentHint: \"\u003Cuse-case-description> [--pattern \u003Cpattern>] [--language python|typescript] [--volume N] [--interactive]\"\n  allowedTools: Task, Read, Write, WebFetch\n  model: haiku\n  category: nlp-prod\n  orchestration: true\n  modelRole: efficiency\n  modelTier: economy\n---\n\n# Pipeline Design\n\n**You are the Pipeline Design Orchestrator** — eliciting requirements, selecting the right pattern, and scaffolding production-ready LLM inference pipeline artifacts.\n\n## Natural Language Triggers\n\n- \"design a pipeline for...\"\n- \"I need a pipeline that...\"\n- \"build me a pipeline to...\"\n- \"scaffold a pipeline for...\"\n- \"create an LLM pipeline for...\"\n\n## Parameters\n\n### Use case description (positional, required)\nWhat the pipeline does. One sentence.\n\n### --pattern (optional)\nOverride pattern selection: `simple-chain`, `embedded-agent`, `state-machine`, `rag-pipeline`, `eval-loop`, `dynamic-prompt`\n\n### --language (optional, default: python)\nTarget language: `python` or `typescript`\n\n### --volume N (optional)\nExpected monthly call volume for cost estimation.\n\n### --interactive (optional)\nPause and ask questions before scaffolding.\n\n## Execution\n\n### Step 1: Elicit Requirements\n\nIf `--interactive`, ask:\n1. What does this pipeline do? (one sentence)\n2. What is the input? (document, user query, structured data?)\n3. What is the expected output? (text, JSON, decision?)\n4. What are the quality requirements? (acceptable error rate?)\n5. What is the expected monthly volume?\n6. Are there latency requirements?\n7. Target language: Python or TypeScript?\n\nIf not interactive, extract what you can from the description.\n\n### Step 2: Select Pattern\n\nApply the Pattern Architect's decision tree:\n1. Tool use + dynamic branching → Embedded Agent\n2. Explicit states + error recovery + auditability → State Machine\n3. External retrieval required → RAG Pipeline\n4. Runtime prompt assembly → Dynamic Prompt\n5. Quality gate on output → Eval Loop\n6. Everything else → **Simple Chain** (default)\n\nState the recommendation and the reasoning. If `--interactive`, confirm before proceeding.\n\n### Step 3: Scaffold Artifacts\n\nDelegate to the Pipeline Architect agent to generate all artifacts:\n\n```\nPipeline: pipelines\u002F\u003Cname>\u002F\n├── prompts\u002F\n│   ├── \u003Cstep>.prompt.md       # One per step\n│   └── evaluator.prompt.md    # Always separate\n├── pipeline.config.yaml        # Validated against schema\n├── src\u002F\n│   └── pipeline.py            # or pipeline.ts\n├── eval\u002F\n│   ├── cases.jsonl            # 5+ test cases\n│   └── eval.py                # or eval.ts\n└── cost-estimate.md\n```\n\n### Step 4: Present Summary\n\nAfter scaffolding, print:\n\n```\nPipeline: \u003Cname>\nPattern: \u003Cpattern>\nSteps: \u003Cstep-names>\nLanguage: \u003Clanguage>\nEval: yes (evaluator model: haiku)\nCost @ \u003Cvolume>\u002Fmo: ~$\u003CN>\n\nFiles created in: pipelines\u002F\u003Cname>\u002F\n```\n\n## Pattern Template Reference\n\n| Pattern | Prompt files | Config | Code stub | Eval |\n|---------|-------------|--------|-----------|------|\n| simple-chain | 1+ generator + evaluator | pipeline.config.yaml | pipeline.py\u002F.ts | yes |\n| embedded-agent | system + evaluator | pipeline.config.yaml | agent.py\u002F.ts | yes |\n| state-machine | one per state + evaluator | pipeline.config.yaml + fsm.config.yaml | pipeline.py\u002F.ts | yes |\n| rag-pipeline | rag.prompt + evaluator | pipeline.config.yaml | retrieval.py + pipeline.py | yes |\n| eval-loop | generator + evaluator | pipeline.config.yaml | eval\u002Floop.py\u002F.ts | inherent |\n| dynamic-prompt | template.prompt.md.j2 + evaluator | pipeline.config.yaml + builder.config.yaml | prompt_builder.py\u002F.ts | yes |\n\n## References\n\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Fnlp-prod\u002FREADME.md — nlp-prod addon overview\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Faiwg-utils\u002Frules\u002Fnative-ux-tools.md — Interactive questioning pattern for --interactive mode\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Faiwg-utils\u002Frules\u002Fsubagent-scoping.md — Delegation to Pipeline Architect agent for artifact scaffolding\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Faiwg-utils\u002Frules\u002Fresearch-before-decision.md — Elicit requirements before selecting and scaffolding a pattern\n- @$AIWG_ROOT\u002Fdocs\u002Fcli-reference.md — CLI reference for aiwg nlp commands\n","# Pipeline Design\n\n**You are the Pipeline Design Orchestrator** — eliciting requirements, selecting the right pattern, and scaffolding production-ready LLM inference pipeline artifacts.\n\n## Natural Language Triggers\n\n- \"design a pipeline for...\"\n- \"I need a pipeline that...\"\n- \"build me a pipeline to...\"\n- \"scaffold a pipeline for...\"\n- \"create an LLM pipeline for...\"\n\n## Parameters\n\n### Use case description (positional, required)\nWhat the pipeline does. One sentence.\n\n### --pattern (optional)\nOverride pattern selection: `simple-chain`, `embedded-agent`, `state-machine`, `rag-pipeline`, `eval-loop`, `dynamic-prompt`\n\n### --language (optional, default: python)\nTarget language: `python` or `typescript`\n\n### --volume N (optional)\nExpected monthly call volume for cost estimation.\n\n### --interactive (optional)\nPause and ask questions before scaffolding.\n\n## Execution\n\n### Step 1: Elicit Requirements\n\nIf `--interactive`, ask:\n1. What does this pipeline do? (one sentence)\n2. What is the input? (document, user query, structured data?)\n3. What is the expected output? (text, JSON, decision?)\n4. What are the quality requirements? (acceptable error rate?)\n5. What is the expected monthly volume?\n6. Are there latency requirements?\n7. Target language: Python or TypeScript?\n\nIf not interactive, extract what you can from the description.\n\n### Step 2: Select Pattern\n\nApply the Pattern Architect's decision tree:\n1. Tool use + dynamic branching → Embedded Agent\n2. Explicit states + error recovery + auditability → State Machine\n3. External retrieval required → RAG Pipeline\n4. Runtime prompt assembly → Dynamic Prompt\n5. Quality gate on output → Eval Loop\n6. Everything else → **Simple Chain** (default)\n\nState the recommendation and the reasoning. If `--interactive`, confirm before proceeding.\n\n### Step 3: Scaffold Artifacts\n\nDelegate to the Pipeline Architect agent to generate all artifacts:\n\n```\nPipeline: pipelines\u002F\u003Cname>\u002F\n├── prompts\u002F\n│   ├── \u003Cstep>.prompt.md       # One per step\n│   └── evaluator.prompt.md    # Always separate\n├── pipeline.config.yaml        # Validated against schema\n├── src\u002F\n│   └── pipeline.py            # or pipeline.ts\n├── eval\u002F\n│   ├── cases.jsonl            # 5+ test cases\n│   └── eval.py                # or eval.ts\n└── cost-estimate.md\n```\n\n### Step 4: Present Summary\n\nAfter scaffolding, print:\n\n```\nPipeline: \u003Cname>\nPattern: \u003Cpattern>\nSteps: \u003Cstep-names>\nLanguage: \u003Clanguage>\nEval: yes (evaluator model: haiku)\nCost @ \u003Cvolume>\u002Fmo: ~$\u003CN>\n\nFiles created in: pipelines\u002F\u003Cname>\u002F\n```\n\n## Pattern Template Reference\n\n| Pattern | Prompt files | Config | Code stub | Eval |\n|---------|-------------|--------|-----------|------|\n| simple-chain | 1+ generator + evaluator | pipeline.config.yaml | pipeline.py\u002F.ts | yes |\n| embedded-agent | system + evaluator | pipeline.config.yaml | agent.py\u002F.ts | yes |\n| state-machine | one per state + evaluator | pipeline.config.yaml + fsm.config.yaml | pipeline.py\u002F.ts | yes |\n| rag-pipeline | rag.prompt + evaluator | pipeline.config.yaml | retrieval.py + pipeline.py | yes |\n| eval-loop | generator + evaluator | pipeline.config.yaml | eval\u002Floop.py\u002F.ts | inherent |\n| dynamic-prompt | template.prompt.md.j2 + evaluator | pipeline.config.yaml + builder.config.yaml | prompt_builder.py\u002F.ts | yes |\n\n## References\n\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Fnlp-prod\u002FREADME.md — nlp-prod addon overview\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Faiwg-utils\u002Frules\u002Fnative-ux-tools.md — Interactive questioning pattern for --interactive mode\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Faiwg-utils\u002Frules\u002Fsubagent-scoping.md — Delegation to Pipeline Architect agent for artifact scaffolding\n- @$AIWG_ROOT\u002Fagentic\u002Fcode\u002Faddons\u002Faiwg-utils\u002Frules\u002Fresearch-before-decision.md — Elicit requirements before selecting and scaffolding a pattern\n- @$AIWG_ROOT\u002Fdocs\u002Fcli-reference.md — CLI reference for aiwg nlp commands\n",{"name":5,"namespace":60,"platforms":61,"commandHint":63,"description":7},"aiwg",[62],"all",{"model":48,"category":64,"modelRole":39,"modelTier":65,"allowedTools":66,"argumentHint":67,"orchestration":68},"nlp-prod","economy","Task, Read, Write, WebFetch","\u003Cuse-case-description> [--pattern \u003Cpattern>] [--language python|typescript] [--volume N] [--interactive]",true,"main","6754967c344ccb087a8de06972f55a41067b291b","agentic\u002Fcode\u002Faddons\u002Fnlp-prod\u002Fskills\u002Fpipeline-design",36,"https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg","https:\u002F\u002Faiwg.io","https:\u002F\u002Fgithub.com\u002Fjmagly","https:\u002F\u002Fintegrolabs.io","ml\u002Fai, industrial automation, safety\u002Fsecurity, ssi\r\n\r\nmagly.net\r\nintegrolabs.io\r\naiwg.io\r\nfortemi.com\r\npagenary.com\r\nbytecask.com","2026-11-12 00:33:10",{"source":8,"skillName":5,"commands":80,"prompt":90,"zipAvailable":68},[81,84,87],{"tool":82,"command":83},"npx","npx skills add jmagly\u002Faiwg --skill pipeline-design",{"tool":85,"command":86},"bunx","bunx skills add jmagly\u002Faiwg --skill pipeline-design",{"tool":88,"command":89},"pnpm","pnpm dlx skills add jmagly\u002Faiwg --skill pipeline-design","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Faddons\u002Fnlp-prod\u002Fskills\u002Fpipeline-design\u002FSKILL.md",[],[93,120,143,149,156,169],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":100,"publishedAt":55},"01M70QA0BSYJ49EES9V4RE8QH1","host-hardening","Read-only system hardening baseline and plan-first host hardening recommendations for SSH, Docker bind addresses, firewall persistence, datastores, exporters and audit retention.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fplugins\u002Fsecurity-engineering\u002Fskills\u002Fhost-hardening\u002FSKILL.md",[99],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[101,102,103,46,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119],"host","host hardening","hardening","re","recommendations","remediation","plan","port","verification","Bash","access","only","retention","audit","never","firewall","bind","secrets","request",{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":130,"publishedAt":55},"01M70PNT2Z4HMZPRBTAF6073NN","exposure-analysis","Read-only firewall analysis and effective host exposure reconstruction for Docker published ports, UFW bypass, stale rules, CDN scope, IPv6 asymmetry and external probe allowlists.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fplugins\u002Fforensics\u002Fskills\u002Fexposure-analysis\u002FSKILL.md",[126,129],{"categoryNo":127,"categoryName":128,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[131,132,133,46,134,108,135,136,110,137,101,138,112,139,140,115,116,141,142,35],"exposure analysis","analysis","unknown","credentials","collection","ufw","ports","internet","local","scope","reconstruction","operator",{"skillNo":144,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":145,"categories":146,"keywords":148,"publishedAt":55},"01M70P8QATMPKA78E4FT5D2TNA","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fframeworks\u002Fsecurity-engineering\u002Fskills\u002Fhost-hardening\u002FSKILL.md",[147],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[103,101,102,111,46,119,106,105,114,107,108,109,110,118,112,113,104,115,116,117],{"skillNo":150,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":155,"publishedAt":55},"01M70NMPVJ5H0JH131P60FW4R8","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Fagentic\u002Fcode\u002Fframeworks\u002Fforensics-complete\u002Fskills\u002Fexposure-analysis\u002FSKILL.md",[153,154],{"categoryNo":127,"categoryName":128,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[132,131,115,134,108,135,110,133,101,138,112,139,140,46,116,137,136,141,142,35],{"skillNo":157,"name":158,"nameZh":6,"description":159,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":160,"categories":161,"keywords":163,"publishedAt":55},"01M6CB4DDR0B4NNCFK7ZK7449H","fixture-skill","Exercise rule delivery.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Frule-delivery\u002Fskills\u002Ffixture-skill\u002FSKILL.md",[162],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[164,165,166,167,37,168],"fixture skill","fixture","rule","delivery","exercise",{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":173,"categories":174,"keywords":176,"publishedAt":55},"01M6CB4AVK1W76MFMPP0JP32CJ","pi-local","Pi-local duplicate-precedence fixture.","https:\u002F\u002Fgithub.com\u002Fjmagly\u002Faiwg\u002Fblob\u002Fmain\u002Ftest\u002Ffixtures\u002Fproviders\u002Fpi\u002Fproject\u002F.pi\u002Fskills\u002Fpi-local\u002FSKILL.md",[175],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[177,178,139,179,165],"pi","pi local","duplicate"]