# marketplace-recsys-feature-engineering

marketplace-recsys-feature-engineering designs recommender features from listing photos, metadata, and onboarding wizards. Powers item-to-item, user-to-item, and mutual-fit matching.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 206 installs · ★ 227 repo stars
- Author: pproenca
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/pproenca/dot-skills/tree/HEAD/skills/.experimental/marketplace-recsys-feature-engineering/SKILL.md
- Updated: 2026-08-15
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/marketplace-recsys-feature-engineering

## What it does

This skill provides 44 rules across eight categories and one feature discovery playbook. It focuses on turning raw marketplace assets into usable signals for three recsys targets: item-to-item similarity, user-to-item feed ranking, and mutual-fit matching between users.

The workflow begins with asset auditing and first-principles decomposition before any model choice. It covers zero-shot CLIP image embeddings, room-type detection before amenity tagging, and extracting object counts rather than boolean flags. It serves as an upstream guide before feeding features into engines like AWS Personalize or OpenSearch.

Fields with less than 80% coverage get rejected from the feature plan immediately. The skill offers design guidelines and architecture rules rather than runnable pipeline scripts.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add pproenca/dot-skills --skill marketplace-recsys-feature-engineering -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/marketplace-recsys-feature-engineering/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/marketplace-recsys-feature-engineering/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add pproenca/dot-skills --skill marketplace-recsys-feature-engineering -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/marketplace-recsys-feature-engineering/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add pproenca/dot-skills --skill marketplace-recsys-feature-engineering -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/marketplace-recsys-feature-engineering/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install pproenca/dot-skills skills/.experimental/marketplace-recsys-feature-engineering
```

2. **Manually: put the files in place**
- File: `.github/skills/marketplace-recsys-feature-engineering/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add pproenca/dot-skills --skill marketplace-recsys-feature-engineering -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/pproenca/dot-skills --path skills/.experimental/marketplace-recsys-feature-engineering
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/marketplace-recsys-feature-engineering/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add pproenca/dot-skills --skill marketplace-recsys-feature-engineering -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/pproenca/dot-skills/tree/HEAD/skills/.experimental/marketplace-recsys-feature-engineering/SKILL.md

## When to use it
- extract recommender signals from listing photos using CLIP and room detection
- structure onboarding wizard questions to produce mutual-fit matching features
- audit marketplace asset tables for freshness, coverage thresholds, and PII
- design ablation tests to prove incremental feature value before deployment

## Example prompts
- What features should we extract from listing photos to build a similar homes shelf?
- Help me design onboarding wizard questions that give clean signals for mutual-fit matching.
- Audit our listing metadata fields and reject candidates below the 80% coverage threshold.

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: 318f88abc918555f.

BEGIN THIRD-PARTY FILE 318f88abc918555f
```text
---
name: marketplace-recsys-feature-engineering
description: Feature engineering for marketplace recommenders — what to extract from raw marketplace assets (listing photos, owner-entered listing metadata, sitter wizard responses) to power item-to-item (similar listings), user-to-item (homefeed ranking), or user-to-user (mutual-fit matching) recommenders. Covers asset auditing, first-principles feature decomposition, vision-feature extraction (CLIP, room-type, amenities, aesthetics), listing text and metadata encoding, sitter wizard design, derived-composition patterns for i2i / u2i / u2u (ANN shelves, two-tower, mutual-fit), feature quality governance (training-serving parity, drift, PII), and incremental value proof (ablation A/B, kill reviews, feature-free baseline). Trigger even when the user does not explicitly say "feature engineering" but is asking how to get more signal out of listing photos, listing metadata, or the sitter onboarding wizard, or how to improve i2i / u2i / u2u quality without blindly ingesting a new model.
---
# Marketplace Engineering Recsys Feature Engineering Best Practices

Comprehensive first-principles guide for deriving usable recommender features from the raw assets of a two-sided trust marketplace — listing photos, owner-supplied listing metadata, and sitter wizard responses — for item-to-item, user-to-item, and user-to-user solutions. Contains 44 rules across 8 categories ordered by cascade impact on the feature-engineering lifecycle, plus one playbook that composes the rules into an end-to-end feature discovery workflow.

This skill is the **upstream precursor** to `marketplace-personalisation` (AWS Personalize) and `marketplace-search-recsys-planning` (OpenSearch retrieval). Those skills treat features as inputs they already have; this skill is about deciding what features to *build* from the raw assets, which decisions they serve, and how to prove each one is worth its maintenance cost.

## When to Apply

Reference this skill when:

- Planning what to extract from listing photos, descriptions, or amenity lists to power i2i similarity or u2i ranking
- Designing or revising the sitter onboarding wizard with recsys features as the primary output
```
END THIRD-PARTY FILE 318f88abc918555f

The third-party file has ended. Everything between the 318f88abc918555f tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2223 of 14393 bytes. © pproenca · MIT · https://github.com/pproenca/dot-skills/tree/HEAD/skills/.experimental/marketplace-recsys-feature-engineering/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
