information-retrieval

Community

Debug and improve search relevance

Authorzhang-ming-hui
Version1.0.0
Installs0

System Documentation

What problem does it solve?

It helps you improve weak skill search results by isolating why relevant skills are missed or ranked too low, then applying targeted fixes to tokenization, expansion, weighting, indexing, or ranking.

Core Features & Use Cases

  • Failure-bucket driven debugging: Reproduce a failing query, map the issue to a specific failure cause, and target the smallest change that improves that bucket.
  • Retrieval pipeline iteration: Tune components such as tokenization, query expansion, field weighting, and ranking so the retrieval engine matches expected skills.
  • Evaluation-driven validation: Re-run the relevant evaluation set and compare before/after metrics to avoid regressions.

Quick Start

Ask an AI to reproduce a failing query from your evaluation output, identify the most likely failure cause (tokenization gap, missing expansion, weighting issue, popularity boost dominance, or missing data), then propose the smallest configurable change in the IR code/config and re-run the affected evaluation bucket to confirm metric improvement.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: information-retrieval
Download link: https://github.com/zhang-ming-hui/ackownledge/archive/main.zip#information-retrieval

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.