Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问clear审计未展示

reading-data读取数据

Agent Skill

用于辅助数据整理、表格处理、CSV/Excel 分析、指标计算和图表准备。它适合让 Agent 清洗字段、汇总数据、发现异常、生成统计口径或把分析结果转成可读说明。使用时需要确认数据来源、字段含义和时间范围,避免把样本数据当全量事实;涉及敏感数据、导出文件或批量写回时,应先确认权限和脱敏边界。

总安装

979

周安装

40

GitHub Stars

公开资料未说明

下载量

314
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

MIT

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:reading-data(读取数据)
来源仓库:https://github.com/tracemem/tracemem-skills
仓库路径:skills/reading-data
安装命令:
npx skills add tracemem/tracemem-skills --skill "reading-data"
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。

AgentSkills.tonpx skills
npx skills add tracemem/tracemem-skills --skill "reading-data"

简介

用于解析和清洗 CSV、Excel 等格式的业务数据。

  • 适合计算指标、识别异常值和准备可视化素材。
  • 通过 GitHub 安装,需确认字段含义和数据时间范围。
  • 处理敏感数据时必须先脱敏并获得合规授权。reading-data 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 导出文件前建议抽样核对格式和内容一致性。

SKILL.md

name
reading-data
description
Instructions for reading data from TraceMem memory.

Skill: TraceMem Reading Data (Data Products)

Purpose

This skill teaches how to safely and correctly read data using TraceMem Data Products. Direct database selection or API calls are prohibited; all data access must be mediated by TraceMem.

When to Use

  • When you need to retrieve information (database rows, API results, resource state).
  • When you need to search or query for specific records.

When NOT to Use

  • Do not use for writing data (use decision_write / insert product).
  • Do not try to use standard SQL libraries or HTTP clients for governed data.

Core Rules

  • Data Products Only: You must access data via a named "Data Product".
  • Operation Specificity: A Data Product usually supports only ONE operation. Use a "read" product (e.g., customers_read) for reading.
  • Purpose is Mandatory: You must declare a purpose string for every read. This purpose must be in the product's allowed_purposes list.
  • Least Privilege: Only request the data you strictly need.

Correct Usage Pattern

  1. Identify the Product:

Use products_list to find the correct Data Product for your need (e.g., orders_read_v1). Check its schema and allowed_purposes.

  1. Execute Read:

Call decision_read with: - decision_id: Your current open decision. - product: The name of the data product. - purpose: A valid purpose string (e.g., verification). - query: The key-value filter (e.g., {"user_id": "123"}).

  1. Handle Results:

The response contains records (list) and summary.

*Example*:

   {
     "records": [{"id": 1, "status": "active"}],
     "summary": {"rows": 1}
   }

Common Mistakes

  • Implicit Purpose: Failing to provide a purpose or guessing one not in the allowed list.
  • Wrong Product: Trying to read from a product designed for insert.
  • Unbounded Queries: Querying without filters (TraceMem may limit result sets by default).

Safety Notes

  • Audit Trails: Your query parameters and the summary of what you read are recorded in the immutable trace.
  • Sensitive Data: Avoid putting PII in the *query* parameters if possible, unless the Data Product is explicitly designed for PII lookups.
  • Read-Before-Write: Always read the current state of an object before attempting to modify it, to ensure your decision is based on fresh data.

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

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

OpenCode

27.37%
按下载量换算86

Antigravity

24.7%
按下载量换算78

Claude Code

19.17%
按下载量换算60

Gemini CLI

11.58%
按下载量换算36

Cursor

7.5%
按下载量换算24

windsurf

3.41%
按下载量换算11

安全审计

暂无安全审计结果可展示。

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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