Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Feature extraction for marketplace recommenders v1.0.0

Use this resource to extract recommender features from a marketplace's raw assets

New

  • 371 downloads
  • 123 likes
  • 218 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install marketplace-feature-engineering

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent identifies and builds recommender features from listing photos, listing metadata and profile-wizard answers, against 44 rules. It first audits the assets, separates raw data from derived features and checks quality and privacy. You get features ready for similar-listing or matching recommendations.

Mostly picked by

Updated November 2, 2026

Card for the “Feature extraction for marketplace recommenders” skill

Details

Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • references/: cards on asset auditing, coverage and freshness
  • references/: cards on derived affinities, embedding caches and data rights

Good fit for

  • You want similar listings suggested from photos
  • You must check rights and privacy before extracting features

Not for

    Tags

    • marketplace
    • recommendation
    • features
    • photos
    • privacy

    About

    You say

    “Which features can we pull from rental photos and descriptions to suggest similar places? Start with an asset audit.”

    Card for the “Feature extraction for marketplace recommenders” skill

    Skill published by pproenca under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to extract recommender features from a marketplace's raw assets
    3. Review and adjust the result

    Content preview

    Files

    • AGENTS.md17.1 KB
    • assets/templates/_template.md1.2 KB
    • gotchas.md0.7 KB
    • LICENSE1.1 KB
    • metadata.json4.2 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • README.md7.1 KB
    • references/_sections.md4 KB
    • references/audit-measure-coverage-before-modelling.md2 KB
    • references/audit-quantify-freshness-per-asset.md2.1 KB
    • references/audit-sample-every-asset-type-end-to-end.md2 KB
    • references/audit-separate-raw-assets-from-derived-features.md2 KB
    • references/audit-verify-rights-and-privacy-before-extraction.md2 KB
    • references/derive-cache-user-embedding-with-short-ttl.md2.2 KB
    • references/derive-decompose-affinity-into-interpretable-subscores.md2.3 KB
    • references/derive-fuse-modalities-before-item-similarity.md2.2 KB
    • references/derive-precompute-i2i-nearest-neighbours-offline.md2.2 KB
    • references/derive-score-u2u-as-symmetric-mutual-fit.md2.2 KB
    • references/derive-use-two-tower-for-user-item-affinity.md2.7 KB
    • references/firstp-ask-what-signal-a-human-uses.md2.2 KB
    • references/firstp-kill-features-a-popularity-baseline-already-captures.md2.2 KB
    • references/firstp-prefer-directly-observed-over-learned.md2 KB
    • references/firstp-reject-features-you-cannot-serve-at-inference.md2.3 KB
    • references/firstp-start-from-the-decision-not-the-algorithm.md2.5 KB
    • references/firstp-tie-every-feature-to-a-specific-solution.md1.9 KB
    • references/listing-declare-categorical-fields-for-bounded-vocabularies.md2.5 KB
    • references/listing-embed-description-with-pretrained-sentence-encoder.md2 KB
    • references/listing-encode-pet-requirements-as-structured-triples.md2.2 KB
    • references/listing-extract-stay-duration-shape-not-just-length.md2 KB
    • references/listing-hash-geo-to-hierarchies-not-raw-lat-lon.md1.8 KB
    • references/listing-multi-hot-encode-amenity-lists.md2.2 KB
    • references/playbooks/discovering.md13.2 KB
    • references/prove-dedicate-random-exploration-slice-to-new-features.md2 KB
    • references/prove-kill-features-that-dont-earn-maintenance.md2.2 KB
    • references/prove-measure-lift-against-feature-ablated-variant.md2 KB
    • references/prove-retain-feature-free-baseline-permanently.md2 KB
    • references/prove-ship-one-feature-at-a-time.md2.1 KB
    • references/quality-freeze-feature-schemas-per-model-version.md2.3 KB
    • references/quality-gate-features-on-coverage-and-drift.md2.5 KB
    • references/quality-scrub-pii-before-features-leave-secure-zone.md2.2 KB
    • references/quality-serve-training-and-inference-from-one-store.md2.5 KB
    • references/quality-version-feature-definitions-in-one-registry.md2.3 KB
    • references/vision-detect-room-types-before-detecting-amenities.md2.2 KB
    • references/vision-extract-per-object-counts-not-just-presence.md2.2 KB
    • references/vision-fine-tune-on-your-domain-when-clip-underperforms.md2.2 KB
    • references/vision-pool-embeddings-across-a-listings-photo-set.md2 KB
    • references/vision-quantify-image-quality-separately-from-content.md2.1 KB
    • references/vision-use-clip-for-zero-shot-listing-embeddings.md2 KB
    • references/wizard-capture-experience-as-counts-and-dates.md2.5 KB
    • references/wizard-make-skips-genuine-and-log-them.md2.2 KB
    • references/wizard-order-questions-by-information-gain.md2.3 KB
    • references/wizard-prefer-multiple-choice-over-free-text.md2.3 KB
    • references/wizard-separate-hard-constraints-from-soft-preferences.md2.4 KB
    • SKILL.md14.2 KB

    Origin and licence

    Listed unmodified

    Author
    pproenca
    Source
    github.com/pproenca/dot-skills/tree/cf93c57cac89d6fc3e4194686000411567f5caf3/skills/.experimental/marketplace-recsys-feature-engineering
    Licence
    MIT

    Notice

    « marketplace-recsys-feature-engineering » (pproenca, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.