Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问许可证需确认审计通过

rpi-researchRPI 研究

Agent Skill

rpi-research 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

198

周安装

8

GitHub Stars

1

下载量

62
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:rpi-research(RPI 研究)
来源仓库:https://github.com/avoidthekitchen/agent-agnostic-skills
仓库路径:skills/rpi-research
安装命令:
npx skills add https://github.com/avoidthekitchen/agent-agnostic-skills --skill rpi-research
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/avoidthekitchen/agent-agnostic-skills --skill rpi-research

简介

rpi-research 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词快速定位候选结果时使用。

  • 适用于研究检索类任务,可结合来源仓库和原始 README 核验具体用法。
  • 通过 npx skills add 命令从 GitHub 安装,需确认权限范围和维护状态。
  • 使用前建议检查是否会触发联网、命令执行或文件读写等敏感操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

RPI Research

Use this skill to answer complex codebase questions by combining targeted file reading, parallel investigation, and synthesis into a saved research note.

Initial Response

When the skill is invoked, respond with:

I'm ready to research the codebase. Please provide your research question or area of interest, and I'll analyze it thoroughly by exploring relevant components and connections.

Then wait for the user's research query.

Workflow After Receiving the Query

  1. Read directly referenced files first.
  • If the user names specific files, read those files fully before delegating work.
  • Read in the main context first so sub-investigations start from accurate context.
  1. Decompose the question into research tracks.
  • Break the request into clear research areas (components, flows, patterns, ownership boundaries).
  • Track subtasks in a task list so coverage is visible and no branch is dropped.
  • Map likely directories/files before parallel execution.
  1. Run parallel sub-investigations.
  • Spawn multiple focused sub-agents/tasks to research different tracks concurrently.
  • Keep each sub-task narrow and concrete (for example: API entrypoints, data model usage, or error handling path).
  • If sub-agents are unavailable, execute the same tracks sequentially and keep notes separated by track.
  1. Synthesize findings after all tracks finish.
  • Wait for all active tracks to complete before writing conclusions.
  • Reconcile overlaps and conflicts across tracks.
  • Capture concrete file references with 1-based line numbers.
  • Highlight cross-component relationships and architectural decisions.
  1. Produce the research memo in this structure.
---
date: [Current date and time in ISO format]
researcher: [User's Name] / [AI Agent Name]
topic: "[User's Question/Topic]"
tags: [research, codebase, relevant-component-names]
status: complete
---

# Research: [User's Question/Topic]

## Research Question
[Original user query]

## Summary
[Direct answer with high-signal findings]

## Detailed Findings

### [Component/Area 1]
- Finding with citation (`path/to/file.ext:line`)
- Why it matters to the question
- Connection to other areas

### [Component/Area 2]
- Finding with citation (`path/to/file.ext:line`)
- Why it matters to the question
- Connection to other areas

## Code References
- `path/to/file.ext:line` - What is relevant there
- `path/to/other.ext:line` - Why it supports the conclusion

## Architecture Insights
[Design patterns, conventions, and notable tradeoffs]

## Open Questions
[Remaining unknowns and how to resolve them]
  1. Save and present the result.
  • Save research notes to rpi/research/TIMESTAMP_research_topic.md.

- For TIMESTAMP, Use a Windows-safe timestamp for TIMESTAMP: YYYY-MM-DD-HH-MM (e.g., 2026-03-05-22-10).

  • Use a short snake_case topic slug for the filename suffix topic.
  • Reply to the user with a concise summary plus key file references.

Quality Bar

  • Prefer evidence over speculation; when inferring, label it clearly.
  • Cite enough references that another engineer can quickly verify conclusions.
  • Keep the memo self-contained so it is useful without reading raw task logs.
  • Prioritize high-signal findings over exhaustive but low-value dumps.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.54%
按下载量换算21

Claude

31.37%
按下载量换算19

Cursor

16.11%
按下载量换算10

Gemini CLI

9.8%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/avoidthekitchen/agent-agnostic-skills --skill rpi-research 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills