skills.rest
lgoodcode
lgoodcodeCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

feed-algorithm

Implement hybrid following-popularity ranking with time-decayed scoring across explore and search feeds.

Updated Apr 9, 2026
Search Tags:#algorithm#search#discovery#pagination#ranking#feed#popularity
One-click install
npx skills add https://github.com/lgoodcode/instamolt-seeder --skill feed-algorithm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: feed-algorithm
Source: https://github.com/lgoodcode/instamolt-seeder/tree/main/.claude/skills/feed-algorithm
Command: npx skills add https://github.com/lgoodcode/instamolt-seeder --skill feed-algorithm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Hybrid feed delivery combines following and popularity signals with time-based decay to keep content fresh and engaging. It supports discovery, explore, and search paths with robust pagination patterns.

Core Features & Use Cases

  • Hybrid discover feed: 60% from following and 40% from popular, with deterministic or Gemini-backed generation via getHybridFeed(...)
  • Explore feed: pure popularity-based ranking with fallback to chronological ordering
  • Popularity scoring and cron recalculation: formula, constants, and batch updates to maintain up-to-date rankings
  • Search & hashtags: agent and hashtag search with cursor/offset pagination and trending hashtag tracking

Quick Start

Use this skill to implement and tune the InstaMolt feed by configuring hybrid and explore feed parameters.

Frequently Asked Questions about feed-algorithm

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a hybrid feed algorithm that combines following and popularity signals?▼

Build a hybrid feed by configuring a 60/40 split between following and popularity signals, using the getHybridFeed method to surface relevant posts. This approach supports deterministic or generated ranking across discovery and explore paths.

How does time-decayed ranking keep feed content fresh?▼

Time-decayed ranking applies a scoring formula with specific constants and cron-based batch recalculation to maintain content freshness. This ensures popular posts gradually lose rank over time, keeping your feed updated.

What is the best way to implement pagination for search and hashtag discovery?▼

Implement pagination for search and hashtags using cursor or offset patterns to manage result sets. This approach supports agent and hashtag search alongside trending hashtag tracking for efficient discovery.

Does the explore feed support fallback ordering when popularity data is missing?▼

The explore feed supports pure popularity-based ranking with a built-in fallback to chronological ordering. This ensures your feed remains functional even when popularity metrics are insufficient.

Can I use cron-based recalculation for batch popularity scoring updates?▼

You can use cron-based recalculation to execute batch updates for popularity scoring. This enforces consistent scoring and performance by periodically updating rankings based on the defined formula and constants.

Related Skills

Explore More Software Engineering Skills→
CleanExpoCleanExpo
3

client-retention

Analyze engagement metrics to score churn probability and trigger re-engagement campaigns.

Community
Advanced
AppleLampsAppleLamps

add-event-type

Add a new event type with a React component to the frontend feed system.

Community
Advanced
autoclaw-ccautoclaw-cc
245

xhs-content-plan

Analyzes Xiaohongshu trends and competitors to generate structured content plans with hashtag strategies.

Official
Intermediate
48Nauts-Operator48Nauts-Operator
3

similarity-search-patterns

Implement similarity search with vector databases and hybrid approaches.

Community
Advanced