add-rerank-preset

Define custom reranking presets for MCP search tools with scoring weights and overlay masks.

11|3|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/artk0de/TeaRAGs-MCP --skill add-rerank-preset
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: add-rerank-preset
Source: https://github.com/artk0de/TeaRAGs-MCP/tree/main/.claude/skills/add-rerank-preset
Command: npx skills add https://github.com/artk0de/TeaRAGs-MCP --skill add-rerank-preset

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the challenge of generic code search results by allowing developers to define custom ranking strategies that prioritize stable, battle-tested, or specific domain-relevant code over simple semantic matches.

Core Features & Use Cases

  • Custom Scoring Weights: Define how much influence specific signals like churn, authorship, or bug-fix rates have on search results.
  • Overlay Masking: Curate exactly which metadata signals are displayed to the agent during ranking.
  • Use Case: Create a security-audit preset that heavily penalizes code with high bug-fix rates and recent churn, ensuring the agent prioritizes stable, secure modules during vulnerability assessments.

Quick Start

Create a new rerank preset file in the appropriate trajectory directory and register it in the barrel index to make it available to your MCP tools.

Frequently Asked Questions about add-rerank-preset

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

FAQPage Schema
How do I customize code search ranking logic for MCP tools?

To customize code search ranking for MCP tools, you implement a TypeScript RerankPreset interface that defines scoring weights and overlay masks to prioritize specific code chunks in trajectory-based retrieval.

How does custom reranking improve trajectory-based code search results?

Custom reranking improves trajectory-based code search by applying scoring weights to signals like churn or bug-fix rates, allowing you to prioritize stable, battle-tested, or domain-specific code over simple semantic matches.

Can I prioritize stable code during security audits using MCP search?

Yes, you can create a security-audit preset that applies scoring weights to penalize code with high bug-fix rates and recent churn, ensuring the agent prioritizes stable, secure modules during vulnerability assessments.

How do I register a custom rerank preset for an MCP search tool?

To register a custom rerank preset, create a new preset file in the appropriate trajectory directory and register it within the TrajectoryRegistry barrel index to make it available to your MCP tools.

What metadata signals can I mask during MCP code retrieval?

You can use overlay masking to curate exactly which metadata signals, such as authorship, churn, or bug-fix rates, are displayed to the agent during the ranking process for code retrieval.

Do I need TypeScript to implement custom scoring weights for MCP search?

Yes, TypeScript is required to implement the RerankPreset interface, which involves configuring scoring weights and overlay masks for trajectory-based code search optimization.