mcpField
C
creator-ranking-quality-review

creator-ranking-quality-review

github.com/sylphxaiData & databases145 installsUpdated today
Repository

About creator-ranking-quality-review

A skill for turning creator-ranking risks in marketplaces into review artifacts with metrics, owners, safeguards, and recovery plans. It covers ranking quality, fairness, abuse, and rollout governance.

Key features

  1. 01Maps marketplace context and ranking signals
  2. 02Separates retrieval, eligibility, and ranking stages
  3. 03Defines metrics, fairness checks, and abuse controls
  4. 04Produces scorecards, decision tables, and event schemas

Use cases

  1. 01Reviewing quality rules for marketplace rankings
  2. 02Planning safeguards against ranking manipulation
  3. 03Preparing rollback and creator feedback processes

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/sylphxai/skills --skill creator-ranking-quality-review

After installing, type/creator-ranking-quality-reviewin Claude Code to confirm it loaded.

Bundled filesagents/openai.yamlreferences/creator-ranking-quality-patterns.md

Details

From the project README

Creator Ranking Quality Review

Use this skill to convert a risky product, operations, trust, or marketplace question into a concrete artifact with owners, gates, metrics, and recovery paths.

Workflow

  1. Identify marketplace surface, user intent, creator incentives, ranking inventory, quality signals, abuse vectors, and business constraints.
  2. Read references/creator-ranking-quality-patterns.md.
  3. Separate retrieval, eligibility, ranking, merchandising, paid placement, review quality, and moderation actions.
  4. Define quality metrics, fairness checks, creator feedback, abuse controls, and experiment/rollback plan.
  5. Produce ranking scorecard, signal matrix, decision table, event schema, and governance cadence.

When not to use

  • Do not use when the job belongs to marketplace-seller-performance-review — Seller KPIs, interventions, appeals. Absorbed quality-scoring patterns.
  • Do not use for generic advice the base model already handles without this skill's specific artifact contract.

Guardrails

  • Do not let revenue or paid promotion silently masquerade as organic quality.
  • Do not rank only by installs/downloads when this creates rich-get-richer loops or spam incentives.
  • Do not expose enough ranking mechanics to enable easy gaming, but give creators actionable quality feedback.
  • Ranking changes need trust, moderation, support, and creator-communication readiness.

Output format

Ranking context:
Marketplace / user intent / creator incentives:

Signal matrix:
| Signal | Purpose | Abuse risk | Weight direction | Guardrail |
| --- | --- | --- | --- | --- |

Ranking decision table:
- <scenario> -> <eligibility/ranking/merchandising/action>

Governance:
- <metric, review cadence, creator/support comms>

Comments

Markdown supported