[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M730S3M4EZ706WZCJR2MGSAC":3,"skill-same:01M730S3M4EZ706WZCJR2MGSAC":48},{"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":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M730S3M4EZ706WZCJR2MGSAC","recsys-pipeline-architect",null,"Design composable recommendation, ranking, and feed pipelines using the six-stage Source→Hydrator→Filter→Scorer→Selector→SideEffect framework popularized by xAI's open-sourced For You algorithm. Use this skill whenever the user is building any system that picks \"the top K items for a (user, context)\" — social feeds, content CMSs, RAG rerankers, task prioritizers, notification triage, search reranking, ad ranking.","SriramGandhiS\u002FSkills","SriramGandhiS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F211849986?v=4","01M72WQCW12BYX71C6WN584TTA","Sriram S","01M72WQCWTJ2RD0ET1X31AF1RC",1,"MIT","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fskills\u002Feverything-claude-code\u002Fskills\u002Frecsys-pipeline-architect\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: recsys-pipeline-architect\ndescription: Design composable recommendation, ranking, and feed pipelines using the six-stage Source→Hydrator→Filter→Scorer→Selector→SideEffect framework popularized by xAI's open-sourced For You algorithm. Use this skill whenever the user is building any system that picks \"the top K items for a (user, context)\" — social feeds, content CMSs, RAG rerankers, task prioritizers, notification triage, search reranking, ad ranking.\norigin: community\n---\n\n# recsys-pipeline-architect\n\nA spec-and-scaffold skill for building composable recommendation, ranking, and feed pipelines. It encodes the **six-stage pattern** — Source → Hydrator → Filter → Scorer → Selector → SideEffect — popularized by xAI's open-sourced [For You algorithm](https:\u002F\u002Fgithub.com\u002Fxai-org\u002Fx-algorithm) (Apache 2.0). This skill is an independent reimplementation of the pattern (MIT) — no code copied from the original.\n\nUpstream: https:\u002F\u002Fgithub.com\u002Fmturac\u002Frecsys-pipeline-architect\n\n## When to Use\n\n- User wants to build any system that picks \"the top K items for a user\u002Fcontext\"\n- User asks \"how should I rank X\" or describes a feed\u002Fpersonalization problem\n- User has a scoring function and needs the pipeline plumbing around it\n- User wants to migrate from a single relevance score to multi-action prediction with tunable weights\n- User is wrapping an LLM\u002FML scorer and needs filters, hydrators, side-effects, and a runnable scaffold in their stack (TypeScript \u002F Go \u002F Python)\n- Triggers: \"recommendation system\", \"feed algorithm\", \"ranking pipeline\", \"for you feed\", \"candidate pipeline\", \"content recommender\", \"pipeline architecture for recsys\", \"RAG retrieval reranker\"\n\n## When NOT to Use\n\n- Model architecture work (transformer design, two-tower retrieval, embedding training) — this skill is plumbing *around* the model, not the model itself\n- Pure ML training pipelines — the scoring function is the user's responsibility\n- Operating a deployed pipeline (monitoring, autoscaling) — out of scope\n\n## The six-stage framework\n\n| # | Stage | Job | Parallel? |\n|---|---|---|---|\n| 1 | **Source** | Fetch candidates from one or more origins | Yes — multiple sources run in parallel |\n| 2 | **Hydrator** | Enrich each candidate with metadata needed for filtering and scoring | Yes — independent hydrators run in parallel |\n| 3 | **Filter** | Drop candidates that should never be shown (blocked, expired, duplicate, ineligible) | Sequential — each filter sees fewer items |\n| 4 | **Scorer** | Assign each surviving candidate one or more scores | Sequential — later scorers see earlier scores |\n| 5 | **Selector** | Sort by final score, return top K | Single op |\n| 6 | **SideEffect** | Cache served IDs, log impressions, emit events, update counters | Async — must never block the response |\n\n### Why this exact order\n\n- Sources before hydration: know what candidates exist before paying to enrich them\n- Hydration before filtering: many filters need metadata the source did not provide\n- Filtering before scoring: scoring is the expensive stage; drop the ineligible first\n- Scorer chain (not single scorer): real systems compose ML scoring + diversity reranking + business rules\n- Selector after scoring: keeps scoring deterministic and cacheable\n- SideEffects last and async: side effects must never block the user response\n\n## Workflow when invoked\n\nWalk the user through these eight steps:\n\n1. **Clarify the use case** (one round, three questions): items being ranked? input context? language\u002Fruntime?\n2. **Identify the candidate sources**: usually in-network (followed\u002Fowned\u002Fsubscribed) + out-of-network (ML retrieval \u002F trending \u002F similar-to-liked)\n3. **List required hydrations**: for each filter and scorer, what data does it need that the source did not provide?\n4. **List the filters**: duplicate, self, age, block\u002Fmute, previously-served, eligibility. Order matters — cheap before expensive.\n5. **Design the scorer chain**: primary (ML) → combiner (multi-action with weights) → diversity → business rules\n6. **Selector**: sort descending by final score, take top K (or stratified mix for in-network\u002Fout-of-network)\n7. **SideEffects**: cache served IDs, emit impression events, update counters, log analytics — all fire-and-forget\n8. **Generate the scaffold** in the user's stack\n\n## Key trade-offs to surface (don't default silently)\n\n### 1. Single score vs multi-action prediction\n\n- **Single score**: train one model to predict relevance. To change behavior → retrain.\n- **Multi-action**: predict `P(action)` for many actions (read, like, share, skip, report), combine with weights at serving time. To change behavior → change weights. No retraining.\n\nThe X For You system uses multi-action with both positive and negative weights. Recommend multi-action when the user expects to tune frequently.\n\n### 2. Candidate isolation in scoring\n\n- **Isolated**: each candidate scored independently. Deterministic, cacheable.\n- **Joint**: candidates attend to each other during scoring (e.g., transformer over batch). More expressive but non-deterministic across batches.\n\nDefault to isolation. Joint only when there's a specific reason (e.g., explicit batch-aware diversity).\n\n### 3. Online vs offline\n\n- **Request-time (online)**: pipeline runs on each request. Latency budget: 100–300ms. Default.\n- **Pre-computed (offline batch)**: pipeline runs periodically, results cached. Lower latency, lower freshness.\n- **Hybrid**: candidate retrieval offline, ranking online.\n\n## Hard rules\n\n1. **Do not invent benchmark numbers.** \"How much faster?\" → \"depends on workload, run it yourself.\"\n2. **Attribution discipline.** When the pattern is referenced, attribute as \"popularized by xAI's open-sourced For You algorithm\" \u002F `github.com\u002Fxai-org\u002Fx-algorithm` (Apache 2.0).\n3. **No trademark use.** Do not name the user's artifact \"X-like\" or use \"For You\" branding. Pattern is free; brand is not. Suggested naming: \"candidate pipeline\", \"feed pipeline\", \"ranking pipeline\", \"recsys pipeline\".\n4. **Surface trade-offs.** Multi-action vs single, isolation vs joint, online vs offline — never default silently.\n5. **The generated scaffold must run.** No pseudocode passing as code.\n6. **Filter order matters.** Cheap before expensive. Universal before user-specific.\n7. **Side effects never block.** Wrap in fire-and-forget patterns (goroutines \u002F promises without await \u002F asyncio tasks).\n\n## Anti-Patterns\n\n- Scoring before filtering (wastes compute on candidates that will be dropped anyway)\n- Synchronous side effects (cache writes \u002F impression emits blocking the response)\n- A single \"relevance\" score when the product needs to tune for multiple objectives (engagement vs safety vs diversity vs ads)\n- Joint scoring as default (non-deterministic, harder to cache, doesn't compose with reranking stages)\n- Generating pseudocode \"for illustration\" — the scaffold must actually run\n\n## Upstream contents\n\nThe upstream repository at https:\u002F\u002Fgithub.com\u002Fmturac\u002Frecsys-pipeline-architect ships:\n\n- Full `SKILL.md` with the complete 8-step workflow\n- 5 load-on-demand reference docs: interfaces in 4 languages (TS\u002FGo\u002FPython\u002FRust), multi-action scoring pattern, candidate isolation, filter cookbook (12 patterns), scorer cookbook (weighted sum, MMR, diversity penalty, position debiasing)\n- 3 runnable example scaffolds, every one green on its test suite:\n  - Strapi v5 plugin (TypeScript \u002F Jest — 3\u002F3 pass)\n  - Zentra-compatible pipeline (Go with generics — 3\u002F3 pass)\n  - PMAI task prioritizer (Python \u002F FastAPI \u002F pytest — 3\u002F3 pass)\n- v0.1.0 release tagged\n- MIT license; pattern attributed to xAI X For You algorithm (Apache 2.0)\n\nInstall via skills.sh: `npx skills add mturac\u002Frecsys-pipeline-architect`\n","# recsys-pipeline-architect\n\nA spec-and-scaffold skill for building composable recommendation, ranking, and feed pipelines. It encodes the **six-stage pattern** — Source → Hydrator → Filter → Scorer → Selector → SideEffect — popularized by xAI's open-sourced [For You algorithm](https:\u002F\u002Fgithub.com\u002Fxai-org\u002Fx-algorithm) (Apache 2.0). This skill is an independent reimplementation of the pattern (MIT) — no code copied from the original.\n\nUpstream: https:\u002F\u002Fgithub.com\u002Fmturac\u002Frecsys-pipeline-architect\n\n## When to Use\n\n- User wants to build any system that picks \"the top K items for a user\u002Fcontext\"\n- User asks \"how should I rank X\" or describes a feed\u002Fpersonalization problem\n- User has a scoring function and needs the pipeline plumbing around it\n- User wants to migrate from a single relevance score to multi-action prediction with tunable weights\n- User is wrapping an LLM\u002FML scorer and needs filters, hydrators, side-effects, and a runnable scaffold in their stack (TypeScript \u002F Go \u002F Python)\n- Triggers: \"recommendation system\", \"feed algorithm\", \"ranking pipeline\", \"for you feed\", \"candidate pipeline\", \"content recommender\", \"pipeline architecture for recsys\", \"RAG retrieval reranker\"\n\n## When NOT to Use\n\n- Model architecture work (transformer design, two-tower retrieval, embedding training) — this skill is plumbing *around* the model, not the model itself\n- Pure ML training pipelines — the scoring function is the user's responsibility\n- Operating a deployed pipeline (monitoring, autoscaling) — out of scope\n\n## The six-stage framework\n\n| # | Stage | Job | Parallel? |\n|---|---|---|---|\n| 1 | **Source** | Fetch candidates from one or more origins | Yes — multiple sources run in parallel |\n| 2 | **Hydrator** | Enrich each candidate with metadata needed for filtering and scoring | Yes — independent hydrators run in parallel |\n| 3 | **Filter** | Drop candidates that should never be shown (blocked, expired, duplicate, ineligible) | Sequential — each filter sees fewer items |\n| 4 | **Scorer** | Assign each surviving candidate one or more scores | Sequential — later scorers see earlier scores |\n| 5 | **Selector** | Sort by final score, return top K | Single op |\n| 6 | **SideEffect** | Cache served IDs, log impressions, emit events, update counters | Async — must never block the response |\n\n### Why this exact order\n\n- Sources before hydration: know what candidates exist before paying to enrich them\n- Hydration before filtering: many filters need metadata the source did not provide\n- Filtering before scoring: scoring is the expensive stage; drop the ineligible first\n- Scorer chain (not single scorer): real systems compose ML scoring + diversity reranking + business rules\n- Selector after scoring: keeps scoring deterministic and cacheable\n- SideEffects last and async: side effects must never block the user response\n\n## Workflow when invoked\n\nWalk the user through these eight steps:\n\n1. **Clarify the use case** (one round, three questions): items being ranked? input context? language\u002Fruntime?\n2. **Identify the candidate sources**: usually in-network (followed\u002Fowned\u002Fsubscribed) + out-of-network (ML retrieval \u002F trending \u002F similar-to-liked)\n3. **List required hydrations**: for each filter and scorer, what data does it need that the source did not provide?\n4. **List the filters**: duplicate, self, age, block\u002Fmute, previously-served, eligibility. Order matters — cheap before expensive.\n5. **Design the scorer chain**: primary (ML) → combiner (multi-action with weights) → diversity → business rules\n6. **Selector**: sort descending by final score, take top K (or stratified mix for in-network\u002Fout-of-network)\n7. **SideEffects**: cache served IDs, emit impression events, update counters, log analytics — all fire-and-forget\n8. **Generate the scaffold** in the user's stack\n\n## Key trade-offs to surface (don't default silently)\n\n### 1. Single score vs multi-action prediction\n\n- **Single score**: train one model to predict relevance. To change behavior → retrain.\n- **Multi-action**: predict `P(action)` for many actions (read, like, share, skip, report), combine with weights at serving time. To change behavior → change weights. No retraining.\n\nThe X For You system uses multi-action with both positive and negative weights. Recommend multi-action when the user expects to tune frequently.\n\n### 2. Candidate isolation in scoring\n\n- **Isolated**: each candidate scored independently. Deterministic, cacheable.\n- **Joint**: candidates attend to each other during scoring (e.g., transformer over batch). More expressive but non-deterministic across batches.\n\nDefault to isolation. Joint only when there's a specific reason (e.g., explicit batch-aware diversity).\n\n### 3. Online vs offline\n\n- **Request-time (online)**: pipeline runs on each request. Latency budget: 100–300ms. Default.\n- **Pre-computed (offline batch)**: pipeline runs periodically, results cached. Lower latency, lower freshness.\n- **Hybrid**: candidate retrieval offline, ranking online.\n\n## Hard rules\n\n1. **Do not invent benchmark numbers.** \"How much faster?\" → \"depends on workload, run it yourself.\"\n2. **Attribution discipline.** When the pattern is referenced, attribute as \"popularized by xAI's open-sourced For You algorithm\" \u002F `github.com\u002Fxai-org\u002Fx-algorithm` (Apache 2.0).\n3. **No trademark use.** Do not name the user's artifact \"X-like\" or use \"For You\" branding. Pattern is free; brand is not. Suggested naming: \"candidate pipeline\", \"feed pipeline\", \"ranking pipeline\", \"recsys pipeline\".\n4. **Surface trade-offs.** Multi-action vs single, isolation vs joint, online vs offline — never default silently.\n5. **The generated scaffold must run.** No pseudocode passing as code.\n6. **Filter order matters.** Cheap before expensive. Universal before user-specific.\n7. **Side effects never block.** Wrap in fire-and-forget patterns (goroutines \u002F promises without await \u002F asyncio tasks).\n\n## Anti-Patterns\n\n- Scoring before filtering (wastes compute on candidates that will be dropped anyway)\n- Synchronous side effects (cache writes \u002F impression emits blocking the response)\n- A single \"relevance\" score when the product needs to tune for multiple objectives (engagement vs safety vs diversity vs ads)\n- Joint scoring as default (non-deterministic, harder to cache, doesn't compose with reranking stages)\n- Generating pseudocode \"for illustration\" — the scaffold must actually run\n\n## Upstream contents\n\nThe upstream repository at https:\u002F\u002Fgithub.com\u002Fmturac\u002Frecsys-pipeline-architect ships:\n\n- Full `SKILL.md` with the complete 8-step workflow\n- 5 load-on-demand reference docs: interfaces in 4 languages (TS\u002FGo\u002FPython\u002FRust), multi-action scoring pattern, candidate isolation, filter cookbook (12 patterns), scorer cookbook (weighted sum, MMR, diversity penalty, position debiasing)\n- 3 runnable example scaffolds, every one green on its test suite:\n  - Strapi v5 plugin (TypeScript \u002F Jest — 3\u002F3 pass)\n  - Zentra-compatible pipeline (Go with generics — 3\u002F3 pass)\n  - PMAI task prioritizer (Python \u002F FastAPI \u002F pytest — 3\u002F3 pass)\n- v0.1.0 release tagged\n- MIT license; pattern attributed to xAI X For You algorithm (Apache 2.0)\n\nInstall via skills.sh: `npx skills add mturac\u002Frecsys-pipeline-architect`\n",{"name":5,"origin":24,"description":7},"community","master","d6ca01da0f24a3b0e26e4a8b7b66cbd25d429f5d","skills\u002Feverything-claude-code\u002Fskills\u002Frecsys-pipeline-architect",0,"https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills","https:\u002F\u002Fgithub.com\u002FSriramGandhiS","https:\u002F\u002Fsriram.website\u002F","Curious guy who keeps tinkering until something works","2026-05-20 16:21:34",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add SriramGandhiS\u002FSkills --skill recsys-pipeline-architect",{"tool":40,"command":41},"bunx","bunx skills add SriramGandhiS\u002FSkills --skill recsys-pipeline-architect",{"tool":43,"command":44},"pnpm","pnpm dlx skills add SriramGandhiS\u002FSkills --skill recsys-pipeline-architect","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fskills\u002Feverything-claude-code\u002Fskills\u002Frecsys-pipeline-architect\u002FSKILL.md",true,[],[49,56,61,67,74,81],{"skillNo":50,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":53,"categories":54,"keywords":55,"publishedAt":19},"01M732609EPJWV13PVTX3GT4TM","webgpu-threejs-tsl","Comprehensive guide for developing WebGPU-enabled Three.js applications using TSL (Three.js Shading Language). Covers WebGPU renderer setup, TSL syntax and node materials, compute shaders, post-processing effects, and WGSL integration. Use this skill when working with Three.js WebGPU, TSL shaders, node materials, or GPU compute in Three.js.","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fwebgpu-threejs-tsl\u002Fskills\u002Fwebgpu-threejs-tsl\u002FSKILL.md",[],[],{"skillNo":57,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7325XM47AGWB6K6C4VXKE6B","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fwebgpu-threejs-tsl\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7325WP6DES69JKC840VMFEP","verification-loop","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fweb-assets\u002Fdocs\u002Fzh-TW\u002Fskills\u002Fverification-loop\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M7325TFQF1DZTQ6J3WGQQ1VR","tdd-workflow","Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fweb-assets\u002Fdocs\u002Fzh-TW\u002Fskills\u002Ftdd-workflow\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7325RGV9AY9PK92D2V8KKPW","strategic-compact","Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fweb-assets\u002Fdocs\u002Fzh-TW\u002Fskills\u002Fstrategic-compact\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7325MZC28SE86E7NMCNEFS4","security-review","Use this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment\u002Fsensitive features. Provides comprehensive security checklist and patterns.","https:\u002F\u002Fgithub.com\u002FSriramGandhiS\u002FSkills\u002Fblob\u002Fmaster\u002Fweb-assets\u002Fdocs\u002Fzh-TW\u002Fskills\u002Fsecurity-review\u002FSKILL.md",[],[]]