跳到技能内容
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
词映 AI导航
工具库Skills 技能
我的收藏
词映 AI导航
发现工具选型指南Skills 技能AI 快讯AI图片AI视频AI文档AI办公AI助手AI搜索AI编程AI音频AI设计AI智能体模型与开发AI学习模型评测AI提示词内容检测AI金融AI 3D扫描与识别其他工具跨境电商 AI
Skills directory/NomaDamas/AutoRAG-Research/create-retrieval-plugin

Codex AutoRAG Retrieval Plugin for Query and Result Contracts

Implement a custom AutoRAG-Research strategy using the repository's text and stored-query entrypoints. Verify IDs, scores, and top_k locally.

中文

Install in Codex

Codex with an AutoRAG-Research checkout, local Python tests, and non-sensitive query and document fixtures.

  1. Open the pinned GitHub original and retain the complete `.agents/skills/create-retrieval-plugin/` directory in the checkout.
  2. Open Codex in that checkout and invoke `$create-retrieval-plugin` with the example task and desired output.

Example task

Give `$create-retrieval-plugin` three fictional documents. Verify that raw text and a stored query ID return the same result shape under matching conditions.

What to check

Both methods expose document IDs and scores. Tests reproduce result limits and empty results.

Original SKILL.md on GitHub · Updated 2026-11-13

Frequently asked question

Should the plugin implement its own concurrency control?

The upstream base pipeline coordinates concurrency. Implement the retrieval methods and their result contract.

Related skills from this repository

  • Codex AutoRAG Ingestor Plugin for Local Dataset Mapping
Browse Codex skills in the Chinese directory
词映 AI 导航Skills 技能
工具分类全部工具已本地化技能选型指南AI 快讯