一键导入
recommendation-engine-builder
Build and extend the recommendation engine. Use when adding related posts, trending, personalized feeds, or saved posts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Build and extend the recommendation engine. Use when adding related posts, trending, personalized feeds, or saved posts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Build and extend the personalization system. Use when adding behavior tracking, user preferences, personalized feeds, or privacy controls.
Build and extend the search system. Use when adding search routes, indexing, ranking, filters, or search analytics.
Seo Keyword Research capability
Affiliate click and conversion tracking skill for DZIRE_v1. Records clicks, attributes conversions via session/click ID, and manages payout records.
Stripe payment integration skill for DZIRE_v1. Handles Checkout sessions, webhook verification, customer management, and payment security.
Revenue analytics and MRR/ARR reporting skill for DZIRE_v1. Aggregates subscription, affiliate, and sponsor revenue into summary metrics.
| name | recommendation-engine-builder |
| description | Build and extend the recommendation engine. Use when adding related posts, trending, personalized feeds, or saved posts. |
| category | build |
| step | 9 |
| version | v1.0 |
| inputs | ["user request","existing backend/app/recommendations/ structure","database/schemas/recommendations.sql","api/contracts/recommendations.json"] |
| outputs | ["Recommendation routes and services","Scoring logic","Saved posts system","Frontend recommendation components"] |
Build and extend the recommendation engine for DZIRE_v1 — related posts, trending, personalized feeds, and saved post collections.
backend/app/recommendations/
├── models.py — Recommendation, SavedPost ORM models
├── schemas.py — Pydantic schemas
├── routes.py — FastAPI route handlers
├── services.py — Business logic (related, trending, for-you, saved posts)
├── scoring.py — Recommendation score computation
└── personalization.py — User-specific personalization score
score = tag_overlap*0.5 + category_match + keyword_similarity + popularity + recency + user_interest
backend/app/main.py under # Step 9 routersdocs/recommendations.md for full documentation